Neural Network Output Feedback Control Research Articles

Steering-Angle Prediction and Controller Design Based on Improved YOLOv5 for Steering-by-Wire System.

A crucial role is played by steering-angle prediction in the control of autonomous vehicles (AVs). It mainly includes the prediction and control of the steering angle. However, the prediction accuracy and calculation efficiency of traditional YOLOv5 are limited. For the control of the steering angle, angular velocity is difficult to measure, and the angle control effect is affected by external disturbances and unknown friction. This paper proposes a lightweight steering angle prediction network model called YOLOv5Ms, based on YOLOv5, aiming to achieve accurate prediction while enhancing computational efficiency. Additionally, an adaptive output feedback control scheme with output constraints based on neural networks is proposed to regulate the predicted steering angle using the YOLOv5Ms algorithm effectively. Firstly, given that most lane-line data sets consist of simulated images and lack diversity, a novel lane data set derived from real roads is manually created to train the proposed network model. To improve real-time accuracy in steering-angle prediction and enhance effectiveness in steering control, we update the bounding box regression loss function with the generalized intersection over union (GIoU) to Shape-IoU_Loss as a better-converging regression loss function for bounding-box improvement. The YOLOv5Ms model achieves a 30.34% reduction in weight storage space while simultaneously improving accuracy by 7.38% compared to the YOLOv5s model. Furthermore, an adaptive output feedback control scheme with output constraints based on neural networks is introduced to regulate the predicted steering angle via YOLOv5Ms effectively. Moreover, utilizing the backstepping control method and introducing the Lyapunov barrier function enables us to design an adaptive neural network output feedback controller with output constraints. Finally, a strict stability analysis based on Lyapunov stability theory ensures the boundedness of all signals within the closed-loop system. Numerical simulations and experiments have shown that the proposed method provides a 39.16% better root mean squared error (RMSE) score than traditional backstepping control, and it achieves good estimation performance for angles, angular velocity, and unknown disturbances.

Distributed Adaptive Optimization Algorithm for High-Order Nonlinear Multi-Agent Stochastic Systems with Lévy Noise.

An adaptive neural network output-feedback control strategy is proposed in this paper for the distributed optimization problem (DOP) of high-order nonlinear stochastic multi-agent systems (MASs) driven by Lévy noise. On the basis of the penalty-function method, the consensus constraint is removed and the global objective function (GOF) is reconstructed. The stability of the system is analyzed by combining the generalized Itô's formula with the Lyapunov function method. Moreover, the command filtering mechanism is introduced to solve the "complexity explosion" problem in the process of designing virtual controller, and the filter errors are compensated by introducing compensating signals. The proposed algorithm has been proved that the outputs of all agents converge to the optimal solution of the DOP with bounded errors. The simulation results demonstrate the effectiveness of the proposed approach.

Event-triggered adaptive neural inverse optimal output-feedback control of steer-by-wire vehicle systems with prescribed performance

In this article, an event-triggered adaptive neural network (NN) output-feedback inverse optimal control issue is investigated for the steer-by-wire vehicle (SBWV) systems. Firstly, NNs are utilized to approximate the unknown nonlinear dynamics and the auxiliary system of SBWV systems is established. Then, a NN state observer is constructed to estimate the unmeasured states. To obtain better tracking performance, the prescribed performance technique is introduced to constrain the tracking error. An event-triggered mechanism (ETM) is established to decrease the numbers of controller execution times. Subsequently, an event-triggered adaptive NN inverse optimal output-feedback control algorithm is proposed by employing the backstepping control theory. It is proved that the developed control method can not only ensure the stability of the SBWV systems, but also guarantee the tracking error does not exceed the prescribed performance bound and converges to a small neighborhood of zero. Finally, simulation results are given to verify the validity of the proposed control method.

Event-Triggered Adaptive Fuzzy Neural Network Output Feedback Control for Constrained Stochastic Nonlinear Systems.

This article investigates the problem of command-filtered event-triggered adaptive fuzzy neural network (FNN) output feedback control for stochastic nonlinear systems (SNSs) with time-varying asymmetric constraints and input saturation. By constructing quartic asymmetric time-varying barrier Lyapunov functions (TVBLFs), all the state variables are not to transgress the prescribed dynamic constraints. The command-filtered backstepping method and the error compensation mechanism are combined to eliminate the issue of "computational explosion" and compensate the filtering errors. An FNN observer is developed to estimate the unmeasured states. The event-triggered mechanism is introduced to improve the efficiency in resource utilization. It is shown that the tracking error can converge to a small neighborhood of the origin, and all signals in the closed-loop systems are bounded. Finally, a physical example is used to verify the feasibility of the theoretical results.

Event-triggering adaptive neural network output feedback control for networked systems under false data injection attacks

This paper investigates the event-triggering adaptive neural network (NN) output feedback control problem for networked control systems subject to false data injection (FDI) attacks. To reduce unnecessary data transmission, an event-triggered mechanism (ETM) related to the measured output is designed on the channel of the sensor-to-observer. Meanwhile, a novel ETM associated with the observer state and parameter estimation signal is devised on the channel of the observer-to-controller to save communication resources on the controller-to-actuator channel. Subsequently, NN technique is employed to approximate FDI attacks, then an adaptive neural output feedback controller is designed to counteract the impact of FDI attacks on the system. Sufficient conditions for the semi-globally uniformly ultimately bounded of the system are obtained by means of the Lyapunov function. Additionally, a co-design scheme for the control gain, observer gain and event-triggered parameters is presented. Finally, a practical example is given to demonstrate the effectiveness of the proposed method.

Adaptive Neural Network Output Feedback Control of Uncertain Underactuated Systems With Actuated and Unactuated State Constraints

Underactuated systems are widely applied in industry, construction, manufacturing, etc., and the complex working environment puts forward higher demands for safety and transient performance. Hence, it is necessary to consider how to simultaneously ensure actuated and unactuated motion constraints by fewer control inputs, especially when systems suffer from model uncertainties, unavailable velocities, etc. Unfortunately, it is still a significant challenge to overcome in real applications and theoretical analysis. Additionally, most existing studies merely consider the specific control objects and few general methods are applicable to a class of underactuated systems. To this end, we design a new adaptive output-feedback controller for a class of uncertain underactuated systems. Compared with existing methods only handling actuated constraints, an important merit of this article is that by introducing the elaborately designed coupling term composed of actuated and unactuated constraints together, all state variables are kept within the preset time-variant ranges and converge to their desired values. Furthermore, a new Lyapunov function candidate is utilized to provide a theoretical guarantee. As far as we know, without the need of exact model knowledge and velocity feedback, this article provides the first solution to achieve accurate motion control and state constraints for both actuated and unactuated variables, which is meaningful both theoretically and practically. Meanwhile, the asymptotic stability of the equilibrium point for the closed-loop system is proven by utilizing Lyapunov techniques and Barbalat’s lemma. For verification, the presented controller is applied to underactuated overhead and rotary cranes, respectively, together with detailed theoretical analysis and experimental validations.

Neural Network Adaptive Output-Feedback Optimal Control for Active Suspension Systems

The adaptive neural network (NN) output-feedback optimal control issue has been investigated for a quarter-car active electric suspension systems, where the suspension stiffness is unknown and partial state variables are unavailable for measurement. NNs are utilized to identify unknown nonlinearities, and an NN state observer is devised to estimate the unmeasurable states. For each backstepping step, via reinforcement learning (RL), a critic–actor architecture is designed to get the approximation solution of Hamilton–Jacobi–Bellman (HJB) equations and actual and virtual optimization controllers are designed, in which the input saturation constraint and road interference are considered. It is analytically proved that all controlled system signals remain bounded, while the power of the control input signal, as well as the amplitude of the vertical displacement, has been minimized. A comparative simulation is eventually given to elaborate the feasibility of the developed control algorithm.

A new adaptive deep neural network controller based on sparse auto‐encoder for the antilock bracking system systems subject to high constraints

AbstractWe contribute in the current paper to extend the universal function approximation property of deep learning hybrid strategy to design both : (a) adaptive deep neural network observer to estimate derivatives of the tracking error dynamics (b) and robust deep neural network (DNN) output feedback control scheme (OFCS) that will overcome successfully effects of both parametric variations and modeling errors. First, the designed controller employs OFCS to linearize the partially known antilock bracking (ABS) dynamics, and then, the dynamic compensator is involved to stabilize the linearized system. However, conventional controllers suffer from limitations due to the presence of these high uncertainties. Therefore, we aim to demonstrate for the first time in the control area the feasibility of applying deep learning algorithm based on sparse auto‐encoder as an approximator for neglected dynamics and uncertain parameters of the ABS. The estimated states are used as inputs to the neural network (NN) and in the adaptation laws as an error signal. Simulations of the proposed control algorithm based adaptive DNN observer are conducted then compared to bang–bang controller, PI controller, and adaptive controller‐based single hidden layer‐neural network with only one‐neuron in hidden layers (SHL(1N)NN) to demonstrate its practical potential. Furthermore, both feasibility and efficiency of involving deep learning algorithm (DLA) in the control area have been successfully confirmed through robustness test.

Observer-Based Adaptive Neural Network Control for Nonstrict-Feedback Nonlinear Systems With Time Delays

An observer-based adaptive neural network control problem was investigated for a class of nonstrict-feedback nonlinear systems with time delays. A state observer was constructed to estimate unknown variables in nonlinear systems. With the approximation ability of RBF NNs and the backstepping technique, an adaptive neural network output feedback control approach was proposed. The designed controller ensures the semi-global uniform boundedness of all signals in the closed-loop system. Finally, the simulation example shows the effectiveness of the proposed control approach.

Observer-based Finite-time Control of Stochastic Non-strict-feedback Nonlinear Systems

This paper investigates the observer-based adaptive finite-time neural control issue of stochastic non-strict-feedback nonlinear systems. By establishing a state observer and utilizing the approximation property of the neural network, an adaptive neural network output-feedback controller is constructed. The controller solves the issue that the states of stochastic nonlinear system cannot be measured, and assures that all signals in the closed-loop system are bounded. Different from the existing adaptive control researches of stochastic nonlinear systems with unmeasured states, the proposed control scheme can guarantee the finite-time stability of the stochastic nonlinear systems. Furthermore, the effectiveness of the proposed control approach is verified by the simulation results.

Adaptive neural network output feedback control of incommensurate fractional-order PMSMs with input saturation via command filtering and state observer

In this paper, an adaptive neural network (NN) output feedback control is investigated for incommensurate fractional-order permanent magnet synchronous motors under the condition of input saturation. First, a NN state observer is presented to obtain the ‘virtual estimate’ of angle speed, where the unknown function is approximated by the NN. Then, in order to solve the input saturation problem, an auxiliary system is developed under fractional-order framework. Next, the command filtered technology with an error compensation mechanism is used to handle the ‘explosion of complexity’ in backstepping and remove the filtering errors. In addition, the frequency distributed model is utilized such that the Lyapunov theory is available in the backstepping design and the system stability is demonstrated. Finally, numerical simulations confirm the availability of the proposed design.

Adaptive neural network output-feedback control of multiple Ackermann steering vehicles formation including motor dynamics with a guaranteed performance

A prescribed performance output-feedback formation control of electrically driven Ackermann steering robotic vehicles is addressed in this paper. Some constraints are imposed to relative distance and angle errors between the cars and a leader. Then, constrained errors are transformed into a new second-order Euler–Lagrange formulation of unconstrained errors via the prescribed performance function technique which inherits all structural properties of the robot dynamics. Based on dynamic surface control design, an observer-based proportional–integral–derivative virtual controller with a prescribed performance is proposed at the first step. Then, an actual controller is proposed at the actuator level to generate input voltage control signals. The proposed controller takes the following advantages: (1) the possible overshoots and controller singularities are avoided based on some predefined performances of relative distance and angle errors, (2) the controller does not require linear and angular velocity measurements, (3) the unknown nonlinearities and exogenous disturbances are effectively compensated by combining a neural network and adaptive robust controller, and (4) the actuator dynamics is compensated in the inner-loop while voltage control signals are directly generated for each robot motors. Lyapunov’s stability analysis proves the stability of the closed-loop control system. Finally, numerical simulation results will show the controller performance.

Observer-based Adaptive Neural Network Output-feedback Control for Nonlinear Strict-feedback Discrete-time Systems

This paper focuses on an observer-based output-feedback controller design for a nonlinear discrete-time system. The major characteristics of this system is that all of the subsystems are in strict-feedback form and all the states of the system are not measurable. An output tracking control problem is firstly considered in this paper. NNs are utilized to approximate unknown functions, while a state observer is designed to approximatethe unvailable states. An adaptive controller is designed on the basis of the backstepping technique. On the basis of the Lyapunov analysis approach, the boundedness of all the signals is provided. The feasibility of the proposed scheme is verified through a simulation example.

Neural-Network Adaptive Output-Feedback Saturation Control for Uncertain Active Suspension Systems.

The adaptive neural-network (NN) output-feedback control problem is investigated for a quarter-car active suspension system. The sprung mass and the suspension stiffness in the considered suspension system are unknown, and the part states are not measured directly. In the control design, NNs are employed to approximate the unknown nonlinear dynamics, and an NN state observer is given to estimate the immeasurable states. By using the adaptive backstepping control design technique and introducing the command filter method, an observer-based NN output-feedback control algorithm is developed, in which the input saturation constraint is compensated via constructing an auxiliary system. It is proved that all the variables of the controlled system are bounded, and the ride comfort, ride safety condition, and suspension space limit are guaranteed. The computer simulation and compared results further show the effectiveness of the proposed control algorithm.

Output Feedback Adaptive Control for Stochastic Non-strict-feedback System with Dead-zone

This paper focuses on the problem of adaptive neural network (NN) control for a class of nonlinear stochastic non-strict feedback system with dead-zone input. A novel adaptive NN output feedback control approach is first proposed for stochastic non-strict feedback nonlinear systems. In order to solve the problem of dead-zone input, a linear decomposition method is proposed. On the basis of the state observer, an output feedback adaptive NN controller is designed by backstepping approach. It is shown that the proposed controller guarantees that all the signals of the closed-loop systems are semi-globally uniformly bounded in probability. Simulation results further illustrate the effectiveness of the proposed approach.

Adaptive neural network output feedback control for stochastic nonlinear systems with full state constraints

This paper presents an adaptive neural network output feedback control method for stochastic nonlinear systems with full state constraints. The barrier Lyapunov functions are used to conquer the effect of state constraints to system performance. The neural network state observer is established to estimate the unmeasured states. By using dynamic surface control technique, the “explosion of complexity” issue existing in the backstepping design is overcome. The proposed control scheme can guarantee that all signals of the system are bounded and the system output can follow the desired signal. Finally, two examples are given to verify the effectiveness of our control method.

Output feedback control for constrained pure-feedback systems: A non-recursive and transformational observer based approach

In this paper we investigate the control problem of uncertain pure-feedback systems under time-varying output constraints using output information only. By making use of the salient cascade properties of pure-feedback systems as well as a novel scaling function, we convert the constrained system into a normal form without constraints. Then by using only one single neural network (NN) unit for nonlinear approximation and one high-gain observer for transformed state estimation, an adaptive NN output feedback control scheme is constructed. Different from existing results in the literature, our method exhibits the following features: (1) achieving semi-global stable control with only output feedback without imposing any additional restrictive condition; (2) avoiding the recursive design procedures required by some typical approaches such as backstepping; and (3) recovering the steady-state tracking performance under the state feedback. Besides, all the signals in the closed-loop are bounded and the output constraints are never violated. The effectiveness and flexibility of the developed method is demonstrated through control design and simulation on the non-trivial aircraft short-period dynamics.

RBF NN observer based adaptive feedback control for the ABS system under parametric uncertainties and modelling errors

An antilock braking (ABS) scheme control is a relatively difficult task due to its uncertain nonlinear dynamics. According to the requirement that the braking process must be fast and robust, we contribute to extending the universal function approximation property of the radial-basis-function (RBF) neural network (NN) to design both: (a) adaptive NN observer to estimate the tracking error dynamics; and (b) intelligent NN output feedback controller (OFC) that will overcome successfully the existing high uncertainties. Notice that the OFC is introduced to linearise the ABS nonlinear system, and the dynamic compensator is involved to stabilise the linearised system. The estimated states are used in the adaptation laws as an error signal. Simulations of the proposed control algorithm based adaptive RBFNN observer are conducted then compared to the Bang-bang controller to demonstrate its practical potential. Furthermore, its efficiency has been successfully confirmed through a robustness test.

Observer-based adaptive neural control for switched stochastic pure-feedback systems with input saturation

The purpose of this paper is to investigate the problem of adaptive neural network (NN) output-feedback tracking control for input saturated switched stochastic nonlinear systems in pure-feedback form with unmeasured states. In order to facilitate the controller design process, a dead zone-based model of saturation is implemented and radial basis function NNs (RBFNNs) are employed to approximate unknown nonlinear functions and to construct an NN switched nonlinear observer to cope with difficulties raised by the presence of immeasurable state variables. Based on the adaptive backstepping technique and Lyapunov function method, an adaptive NN output feedback control scheme is developed. Furthermore, it is proved that the proposed controller can provide that, under arbitrary deterministic switching, all signals in the closed-loop system are semiglobally uniformly ultimately bounded in probability and the tracking error converges to a small neighborhood of the origin. Finally, simulation examples are presented to validate the effectiveness of the proposed adaptive NN control approach.

Distributed cooperative learning for a group of uncertain systems via output feedback and neural networks

This paper studies the problem of adaptive neural network (NN) output-feedback control for a group of uncertain nonlinear multi-agent systems (MASs) from the viewpoint of cooperative learning. It is assumed that all MASs have identical unknown nonlinear dynamic models but carry out different periodic control tasks, i.e., each agent system has its own periodic reference trajectory. By establishing a network topology among systems, we propose a new consensus-based distributed cooperative learning (DCL) law for the unknown weights of radial basis function (RBF) neural networks appearing in output-feedback control laws. The main advantage of such a learning scheme is that all estimated weights converge to a small neighborhood of the optimal value over the union of all system estimated state orbits. Thus, the learned NN weights have better generalization ability than those obtained by traditional NN learning laws. Our control approach also guarantees the convergence of tracking errors and the stability of closed-loop system. Under the assumption that the network topology is undirected and connected, we give a strict proof by verifying the cooperative persisting excitation condition of RBF regression vectors. This condition is defined in our recent work and plays a key role in analyzing the convergence of adaptive parameters. Finally, two simulation examples are provided to verify the effectiveness and advantages of the control scheme proposed in this paper.