Adaptive Radial Basis Function Neural Network Research Articles

Optimized CNN and adaptive RBFNN for channel estimation and hybrid precoding approaches for multi user millimeter wave massive MIMO

ABSTRACT The millimetre-wave (mmWave) communication satisfies the demand for high data rates due to the characteristic of wide bandwidth. Using massive multiple-input multiple-output (MIMO) technology, a significant propagation loss of mmWave communication is effectively compensated. However, it is challenging to provide a specialised radio frequency chain for each antenna due to the constrained physical area with closely spaced antennas and prohibitive power consumption in mmWave massive MIMO systems. This paper presents novel approaches for effective channel estimation and hybrid precoding in mmWave communication systems. To address the challenges of channel estimation, a convolutional neural network (CNN) is utilised, and network parameters are optimised using enhanced whale optimization algorithm (EWOA). The proposed CNN-based channel estimation method aims to accurately estimate the channel in mmWave systems with enhanced efficiency and reduced complexity. By training CNN using EWOA optimisation algorithm, the network parameters are fine-tuned to improve accuracy and generalisation capability of channel estimation process. Furthermore, hybrid precoding is achieved using adaptive radial-basis function neural networks (adaptive RBFNNs) which enables efficient precoding while minimising complexity. Moreover, the adaptive RBFNN approach determines the optimal precoding weights based on channel state information, resulting in a improved performance and a reduced computational overhead. The performance analysis is validated using the MATLAB/Simulink software and offers to provide effectual and reliable mmWave communication systems, facilitating the realisation of high-speed and high-capacity wireless networks.

Adaptive cyber-tolerant finite-time frequency control framework for renewable-integrated power system under deception and periodic denial-of-service attacks

This work concentrates on designing and applying an adaptive cyber-tolerant finite-time frequency control framework for smart power systems under state-dependent sensor deception and periodic denial-of-service (DoS) attacks. An adaptive radial-basis function neural network (RBFNN)-based disturbance observer (DO) is designed to estimate undeniable plant disturbances, which includes structural uncertainties and unmodelled dynamics. Then, the DO-based resilient secondary control law based on the adaptive nonlinear sliding mode strategy for the cyber-physical power system (CPPS) is derived to guarantee global uniform asymptotic stability of the parametric error and the selected nonlinear sliding manifold. Furthermore, an in-depth analysis of uniformly ultimately boundedness (UUB) of the estimation error of RBFNN has been demonstrated. Numerical simulation and qualitative assessment have revealed the proficiency of the applied feedback control framework for the undertaken CPPS and a high-degree of cyber-tolerance against malicious attacks.

Robust trajectory tracking of quadrotors using adaptive radial basis function network compensation control

Radial Basis Function Neural Networks (RBFNN) methods have gained incredible efficiency and applicability in control. This paper presents a nested control strategy for robust trajectory tracking of a quadrotor using adaptive RBF compensation and NN-supervised control embedded with Integrator BackStepping (IBS). The approach addresses the robustness in the presence of modeling uncertainties, sensing noise, and bounded disturbances. The control design is derived from the decentralized inverse dynamics, using adaptive RBFNN for outer-loop disturbance approximation and compensation. In conjunction with an Inner-loop supervised control that stabilizes the quadrotor attitude, preventing initial instability during NN convergence. In addition, an adaptive Extended Kalman Filter (EKF) attenuates noisy signals. Simulation results demonstrate strong adaptability to changes in model parameters, and superior performance when compared to Proportional Integral Derivative (PID), Integrator BackStepping (IBS), and offline decentralized Multi-Layer Perceptron (MLP) algorithms, in terms of parameter convergence, disturbance compensation control, and noise attenuation.

Controlled synchronization of a vibrating screen driven by two motors based on improved sliding mode controlling method.

With a requirement of miniaturization in modern vibrating screens, the vibration synchronization method can no longer meet the process demand, so the controlled synchronization method is introduced in the vibrating screen to achieve zero phase error state and realize the purpose of increasing the amplitude. In this article, the controlled synchronization of a vibrating screen driven by two motors based on improved sliding mode controlling method is investigated. Firstly, according to the theory of mechanical dynamics, the motion state of the vibrating screen is simplified as the electromechanical coupling dynamical model of a vibrating system driven by two inductor motors. And then the synchronization conditions and stability criterion of the vibrating system are derived and numerically analyzed. Based on a master-slave controlling strategy, the controllers of two motors are respectively designed with Super-Twisting sliding mode control (ST-SMC) and backstepping second-order complementary sliding mode control (BSOCSMC), while the uncertainty is estimated by an adaptive radial basis function neural network (ARBFNN). In addition, Lyapunov stability analysis is performed on the two controllers to prove their stability theoretically. Finally, simulation analysis is conducted based on the dynamics model in this paper.

Adaptive sliding mode attitude control of 2-degrees-of-freedom helicopter system with actuator saturation and disturbances

The modelling uncertainties, external disturbance and actuator saturation issues will degrade the performance and even the safety of flight. To improve control performance, this study proposes an adaptive U-model based double sliding control (UDSMC) algorithm combined with a radial basis function neural network (RBFNN) for a nonlinear two-degrees-of-freedom (2-DOF) helicopter system. Firstly, the adaptive RBFNN is designed to approximate the system dynamics with unknown uncertainties. Furthermore, two adaptive laws are designed to deal with unknown external disturbances and actuator saturation errors. The global stability of the proposed helicopter control system is rigorously guaranteed by the Lyapunov stability analysis, realizing precise attitude tracking control. Finally, the comparative experiments with conventional SMC and adaptive SMC algorithms conducted on the Quanser Aero2 platform demonstrate the effectiveness and feasibility of the proposed 2-DOF helicopter control algorithm.

Research on state-parameter estimation of unmanned Tractor—A hybrid method of DEKF and ARBFNN

Unmanned tractor relies on multi-sensor information collection to obtain the current state or parameters. However, when driving in complex field, it will inevitably suffer from the uneven ground, the impact of crop straw and clods et al., which usually poses considerable challenges for multi-sensor to obtain stable and accurate value of states and parameters to realize tractor automatic lane guidance control. Therefore, a novel state and parameter estimation method by mixing the dual extended Kalman filter (DEKF) technology and adaptive radial basis function neural network (ARBFNN) technology is proposed in this paper. Firstly, DEKF technique is applied to estimate key states and initial model time-varying parameters–front/rear axle cornering stiffnesses at the same time. Then, in order to further improve the accuracy of estimation value of front/rear axle cornering stiffnesses during the automatic lane guidance control process, an ARBFNN technology is investigated by taking the heading error, lateral error and initial estimation value of front/rear axle cornering stiffnesses from DEKF as inputs to approach ideal estimation value. Finally, results of automatic lane guidance control scenarios from both simulation and hardware-in-loop (HIL) implementation show that the proposed hybrid estimated method of DEKF and ARBFNN can robustly obtain satisfactory estimation value and automatic lane guidance control performance for tractor when the control system is characterized by both time-varying model parameters and uncertain external energy-bounded disturbance. A comparative study is also conducted to investigate cases to show its effectiveness when the hybrid DEKF-ARBFNN state and parameter estimation method is used and when it is not.

Adaptive neural network based compensation control of quadrotor for robust trajectory tracking

SummaryNeural network (NN) methods have become increasingly efficient and applicable in control. This article presents a novel nested control strategy based on adaptive radial basis function neural networks (RBFNN) and NN supervised control embedded with integrator back stepping (IBS) for a robust position and attitude trajectory tracking of quadrotor aerial robot in presence of modeling uncertainties, sensing noise, and external bounded disturbance. The control philosophy is derived from the decentralized inverse dynamics, where the adaptive RBFNN is adopted for the outer loop to approximate the unknown external disturbance, and adaptively compensate for their effect. In conjunction with a supervised control that stabilizes the attitude in the inner loop by adaptively compensating the unknown and unmodeled uncertainties, avoiding initial instability of the attitude during NN convergence. Furthermore, noise is attenuated by an adaptive extended Kalman filter. Stability analysis was elaborated using Lyapunov stability theory. Simulation of adaptive RBFNN control philosophy proved robustness and effectiveness compared to proportional integral derivative, IBS, and offline decentralized convolution neural network algorithms as it generates the control law by guarantying a fast convergence of parameters, better external disturbance compensation, and noise attenuation. The proposed scheme of robust tracking was validated.

Robust fixed-time H∞ tracking control of UUVs with partial and full state constraints and prescribed performance under input saturation

In this paper, a fixed-time tracking control problem for an unmanned underwater vehicle (UUV) is investigated in the case of strong external sudden disturbances, model uncertainty, time-varying disturbances and specified performance constraints. First, a modified barrier Lyapunov function (QABLF) is proposed to increase applicability by introducing the standard quadratic Lyapunov function (QLF) and the logarithmic asymmetric barrier Lyapunov function (ABLF). The QALBF has two types and can be constructed to avoid violations of full and partial state constraints by parameter modification. Second, adaptive antisaturation appointed-time prescribed performance functions (APPFs) are introduced in the QABLF for the first time to relax the overshoot restriction and reduce the effect of input saturation on the performance constraint. Third, based on the backstepping method, a robust H∞ control strategy is developed to address strong, sudden disturbances. An auxiliary variable is designed to compensate for input saturation. Next, an improved adaptive radial basis function neural network (ARBFNN) is proposed to match the time-varying disturbances and model uncertainty. Finally, composite robust fixed-time control is achieved with high robustness and transient performance, which guarantees that the closed-loop system is fixed-time convergent. Moreover, the feasibility of the proposed controller is verified by Lyapunov analysis and numerical simulation.

Adaptive NN state error PCH trajectory tracking control for unmanned surface vessel with uncertainties and input saturation

AbstractA novel robust state error port controlled Hamiltonian (PCH) trajectory tracking controller of an unmanned surface vessel (USV) subject to time‐varying disturbances, dynamic uncertainties and control input saturation is presented. The proposed control scheme combines the advantages of the high robustness and energy minimization of the state error PCH approach and the approximation capability of adaptive radial basis function neural networks (RBFNNs). Adaptive RBFNNs are used to the time‐varying disturbances of the environment and unknown dynamics uncertainties of the USV model. The state error PCH control approach is designed such that the system can optimize energy consumption, and the state error PCH technique makes the designed trajectory tracking controller be easy to implement in practice. To handle the effect of the control input saturation, a Gaussian error function model is employed. It has been demonstrated that the proposed approach can maintain the USV's trajectory at the desired trajectory, while the closed‐loop control system can guarantee the uniformly ultimate boundedness. The energy consumption model of the USV is constructed to reveal to the energy consumption. Simulation results demonstrate the effectiveness of the proposed controller.

Adaptive neural networks-based integral sliding mode control for T-S fuzzy model of delayed nonlinear systems

This paper presents a method for designing adaptive neural networks (NNs) based on integral sliding mode control (ISMC) for delayed Takagi-Sugeno fuzzy systems with external disturbances. To do this, a fuzzy sliding surface is constructed with the information of the delayed states, and an adaptive radial basis function (RBF) NNs-based fuzzy ISMC is designed. Here, the RBF NN function is utilized to approximate the nonlinear perturbation term. The resulting stability criteria are derived based on suitable Lyapunov functionals with the help of generalized free-weight matrix-based inequality. The resulting conditions ensure the asymptotic stability with H∞ attenuation level. Moreover, the designed sliding motion achieved the reachability condition under the adaptive RBF NNs based on the sliding mode controller. To end with, a design and comparative examples are given to show the success of the nominated control scheme and less conservatism of the derived sufficient conditions.

Research on the Line of Sight Stabilization Control Technology of Optronic Mast under High Oceanic Condition and Big Swaying Movement of Platform

To realize high-performance line of sight (LOS) stabilization control of the optronic mast under high oceanic conditions and big swaying movements of platforms, a composite control method based on an adaptive radial basis function neural network (RBFNN) and sliding mode control (SMC) is proposed. The adaptive RBFNN is used to approximate the nonlinear and parameter-varying ideal model of the optronic mast, so as to compensate for the uncertainties of the system and reduce the big-amplitude chattering phenomenon caused by excessive switching gain in SMC. The adaptive RBFNN is constructed and optimized online based on the state error information in the working process; therefore, no prior training data are required. At the same time, a saturation function is used to replace the sign function for the time-varying hydrodynamic disturbance torque and the friction disturbance torque, which further reduce the chattering phenomenon of the system. The asymptotic stability of the proposed control method has been proven by the Lyapunov stability theory. The applicability of the proposed control method is validated by a series of simulations and experiments.

Hybrid Backstepping Control of a Quadrotor Using a Radial Basis Function Neural Network

This article presents a hybrid backstepping consisting of two robust controllers utilizing the approximation property of a radial basis function neural network (RBFNN) for a quadrotor with time-varying uncertainties. The quadrotor dynamic system is decoupled into two subsystems: the position and the attitude subsystems. As part of the position subsystem, adaptive RBFNN backstepping control (ANNBC) is developed to eliminate the effects of uncertainties, trace the quadrotor’s position, and provide the desired roll and pitch angles commands for the attitude subsystem. Then, adaptive RBFNN backstepping is integrated with integral fast terminal sliding mode control (ANNBIFTSMC) to track the required Euler angles and improve robustness against external disturbances. The proposed technique is advantageous because the quadrotor states trace the reference states in a short period of time without requiring knowledge of dynamic uncertainties and external disturbances. In addition, because the controller gains are based on the desired trajectories, adaptive algorithms are used to update them online. The stability of a closed loop system is proved by Lyapunov theory. Numerical simulations show acceptable attitude and position tracking performances.

Consensus Tracking for High-Order Uncertain Nonlinear MASs via Adaptive Backstepping Approach.

In this article, we focus on the problems of consensus control for nonlinear uncertain multiagent systems (MASs) with both unknown state delays and unknown external disturbances. First, a nonlinear function approximator is proposed for the system uncertainties deriving from unknown nonlinearity for each agent according to adaptive radial basis function neural networks (RBFNNs). By taking advantage of the Lyapunov-Krasovskii functionals (LKFs) approach, we develop a compensation control strategy to eliminate the effects of state delays. Considering the combination of adaptive RBFNNs, LKFs, and backstepping techniques, an adaptive output-feedback approach is raised to construct consensus tracking control protocols and adaptive laws. Then, the proposed consensus tracking scheme can steer the nonlinear MAS synchronizing to the predefined reference signal on account of the Lyapunov stability theory and inequality properties. Finally, simulation results are carried out to verify the validity of the presented theoretical approach.

An Adaptive Inverse Model Control Method of Vehicle Yaw Stability with Active Front Steering Based on Adaptive RBF Neural Networks

In order to enhance vehicle yaw stability, this paper proposes a novel adaptive inverse model control (AIMC) method for active front steering (AFS) design based on adaptive radial basis function (RBF) neural networks. To approximate the uncertain nonlinearity and modeling errors in vehicle dynamics, an adaptive RBF neural network (ARBFNN) controller is firstly designed. For smaller tracking errors and better robustness, a RBF networks-based AIMC system is further established. Apart from the ARBFNN control law, another two RBF neural networks are utilized, which work as model identifier and inverse model controller, respectively. After being trained offline, both of them are updated by using online learning algorithms. For faster learning and stability, adaptive learning rates are developed. Consequently, the inverse model controller is able to generate required steering wheel angle. Finally, the comparative studies are carried out and the simulation results illustrate the robustness and effectiveness of proposed control strategy.

Design of compensatory backstepping controller for nonlinear magnetorheological dampers

• A novel magnetorheological fluid damper inverse model is constructed based on the radial basis function neural network. • The filter performance function is employed in the semi-active suspension system as the reference state. • A unified frame is developed to study the semi-active suspension system with various disturbances. The magnetorheological fluid semi-active suspension system has been widely employed in vehicles. In general, the output force of the magnetorheological fluid damper varies with service time and temperature, which will cause the controller performance to degrade or even fail. To conquer this problem, this study proposes a compensatory backstepping strategy for the magnetorheological fluid suspension system. First, the normalized phenomenological model is introduced to simulate the nonlinear characteristics of the magnetorheological fluid damper. Second, the adaptive radial-basis function neural network is employed to construct the inverse model of the magnetorheological fluid damper. Next, the reference state is calculated by the filter performance function to trade off the vehicle handling stability and ride comfort. Finally, the voltage-force compensator is constructed to improve the robustness of the controller under input time delay, external noise disturbance, and actuator failures. Furthermore, the performances of the magnetorheological fluid damper, inverse magnetorheological fluid model, and compensatory backstepping controller are simulated under bump and random excitations. It is found that the proposed compensatory backstepping controller shows good improvement in the semi-active suspension system with time delay, noise disturbance, and actuator failure.

Research on the method of predicting CEFR core thermal hydraulic parameters based on adaptive radial basis function neural network

Alterations in thermal hydraulic parameters directly affect the safety of reactors. Accurately predicting the trends of key thermal hydraulic parameters under various working conditions can greatly improve reactor safety, thereby effectively preventing the occurrence of nuclear power plant accidents. The thermal hydraulic characteristic parameters in the reactor are affected by many factors, in order to preliminarily study whose forecasting methods and determine the feasibility of neural network forecasting, the China Experimental Fast Reactor (CEFR) is selected as the research target in this study, and the maximum surface temperature of fuel rod sheath and mass flow rate are used as predictive variables. After data samples are generated through the reactor sub channel analysis code (named SUBCHANFLOW), two widely used adaptive neural networks are used to perform the thermal hydraulic parameter forecast analysis of CEFR fuel assembly under steady-state conditions. The 1/2 core model of CEFR is used to perform a single-step predictive analysis of thermal hydraulic parameters under transient conditions. The results show that the adaptive radial basis function (RBF) neural network exhibits a better fitting ability and higher forecasting accuracy than that of the adaptive back propagation neural network, and the maximum error under steady-state conditions is 0.5%. Under transient conditions, poor forecasting accuracy is observed for some local points; however, the adaptive RBF neural network is generally excellent at predicting temperature and mass flow. The mean relative error of temperature does not exceed 1%, and the mean relative error of flow does not exceed 6.5%. The proposed RBF neural network model can provide real-time forecasting in a short time under unstable flow conditions, and its forecasting results have a certain reference value.

Consensus control of multi-agent systems with deception attacks using event-triggered adaptive cognitive control

This paper presents a novel event-triggered adaptive cognitive control to address the consensus problem of multi-agent systems (MASs) with a Leader under deception attacks. By using the reinforcement learning, adaptive radial basis function (RBF) neural networks and sliding mode control, an adaptive cognitive control is developed. This control has two parts: Actor and Critic. The Actor is designed by using adaptive RBF neural networks and sliding mode control, named adaptive sliding mode control. It is used to control the agent. The Critic is constructed utilizing the adaptive RBF neural networks, to evaluate the control performance of the Actor. In addition, to reduce the communication cost, an event triggered mechanism is designed. The Lyapunov stability analysis shows that the proposed event-triggered adaptive cognitive control can ensure the stabilization of the MASs in case of deception attacks. Simulations are performed to validate the feasibility of the proposed event-triggered adaptive cognitive control, indicating that it can decrease the effects of deception attacks and ensure that all Followers can synchronize to the Leader.

Adaptive RBF Neural Network Backstepping Control for Two-Link Robot Manipulators

In this paper, an adaptive radial basis function neural network(RBFNN) backstepping controller is presented for a two-link robot manipulator in the presence of uncertainty and external interferences. RBFNNs are applied to approximate uncertain nonlinear functions, and considering the backstepping technique, an adaptive RBFNN backstepping control strategy is proposed. It is proved by the Lyapunov function that tracking errors converge to a small neighborhood of the equilibrium point and the closed-loop system variables are bounded. The simulation results demonstrate the effectiveness of the presented design method.

Robust backstepping sliding mode aircraft attitude and altitude control based on adaptive neural network using symmetric BLF

In this paper, asymptotic control of attitude and altitude of the aircraft is proposed by employing robust backstepping sliding mode control (BSMC) in conjunction with adaptive radial basis function neural network (RBFNN). Accurate knowledge of the non-linear aerodynamic forces and moments, particularly high-fidelity models, is of paramount importance to arrive at such a control strategy under a continuous dynamic environment. Adaptive RBFNN is used to approximate such an unknown non-linear function by continuously updating the network weights in rapidly varying conditions. Further, adaptation laws are used concurrently with neural networks to update the control power derivatives. These adaptive neural networks are used within the architecture of backstepping, integrated with the sliding surfaces where angular rates act as the virtual controller. The postulation of such law requires only minimal information about the aerodynamic model beyond well-known physical features. Moreover, Barrier Lyapunov Function (BLF) candidate is employed to constrain the state of the plant from transgressing a specific limit. Closed-loop signals are theoretically proved to be semi-globally uniformly ultimately bounded in the sense of Lyapunov. Finally, the robustness of the designed flight control law is explored by appending uncertainties and bounded exogenous disturbances in the plant. The results obtained in the present study signify good control performance where output tracks the reference signals by forcing the system states to remain in the designed sliding surfaces.

Design of Model-Based and Model-Free Robust Control Strategies for Lower Limb Rehabilitation Exoskeletons

Rehabilitation in the form of locomotion assistance and gait training through robotic exoskeletons requires both precision and accuracy to achieve effective results. The essential challenge is to ensure robust tracking of the reference signal, i.e., of the gait or locomotion. This paper presents the design of model-based (MB) and model-free (MF) robust control strategies to achieve desired performance and robustness in terms of transient behavior and steady-state/tracking error, implementable to the locomotion assistance and gait training by exoskeletons. The dynamic responses of the exoskeleton system were investigated with both the control strategies. The study was carried out with a variety of reference signals and performance was evaluated to identify the best suited approach for rehabilitation exoskeletons. In case of the model-based control, a mathematical model of the system was developed using a bond graph modeling technique and a lead compensated H-infinity reference gain controller was designed to ensure robust tracking performance. In the model-free control strategy, however, the system function is approximated using radial basis function neural networks (RBFNNs) and an adaptive proportional-derivative RBFNN controller was designed to achieve the desired results with minimum tracking error. Both strategies make the system robust and stable. However, the MF control strategy is faster for all reference inputs as compared to the MB control strategy i.e., faster to approach the peak value and settle, and rapidly approaches the zero steady-state/tracking error. The rise time in the case of a sinusoidal input for model-free control is 0.4 s faster than the rise time in model-based control. Similarly, the settling time is 3.9 s faster in the case of model-free control, which is a prominent difference and can provide better rehabilitation results.