Finite-Horizon $H_{\infty}$  State Estimation for Periodic Neural Networks Over Fading Channels

Li, Xiaomeng; Zhang, Bin; Li, Panshuo; Zhou, Qi; Lu, Renquan

doi:10.1109/tnnls.2019.2920368

Cited by 77 publications

(44 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been proven that any discrete‐time functions can be approximated by RBF NN in to any arbitrary accuracy over a compact set Ω Z . The unknown function can be expressed as

U false(Z false) = W^{normalT} S false(Z false) + ε false(Z false),

where W and ε ( Z ) are the ideal weight vector and the NN approximation error, respectively, which satisfy

false‖ W false‖ \leq \overset{W}{true}

and

ε false(Z false) \leq \overset{ε}{true}

with

\overset{W}{true} > 0

and

\overset{ε}{true} > 0

being unknown constants.…”

Section: Problem Formulationmentioning

confidence: 99%

Multigradient recursive reinforcement learning NN control for affine nonlinear systems with unmodeled dynamics

Bai

Zhang

Zhou

et al. 2019

Intl J Robust & Nonlinear

Self Cite

View full text Add to dashboard Cite

Summary In this paper, an adaptive reinforcement learning approach is developed for a class of discrete‐time affine nonlinear systems with unmodeled dynamics. The multigradient recursive (MGR) algorithm is employed to solve the local optimal problem, which is inherent in gradient descent method. The MGR radial basis function neural network approximates the utility functions and unmodeled dynamics, which has a faster rate of convergence than that of the gradient descent method. A novel strategic utility function and cost function are defined for the affine systems. Finally, it concludes that all the signals in the closed‐loop system are semiglobal uniformly ultimately bounded through differential Lyapunov function method, and two simulation examples are presented to demonstrate the effectiveness of the proposed scheme.

show abstract

“…It has been proven that any discrete‐time functions can be approximated by RBF NN in to any arbitrary accuracy over a compact set Ω Z . The unknown function can be expressed as

U false(Z false) = W^{normalT} S false(Z false) + ε false(Z false),

where W and ε ( Z ) are the ideal weight vector and the NN approximation error, respectively, which satisfy

false‖ W false‖ \leq \overset{W}{true}

and

ε false(Z false) \leq \overset{ε}{true}

with

\overset{W}{true} > 0

and

\overset{ε}{true} > 0

being unknown constants.…”

Section: Problem Formulationmentioning

confidence: 99%

Multigradient recursive reinforcement learning NN control for affine nonlinear systems with unmodeled dynamics

Bai

Zhang

Zhou

et al. 2019

Intl J Robust & Nonlinear

Self Cite

View full text Add to dashboard Cite

show abstract

“…Adaptive output feedback control was presented for a class of nonlinear systems with quantized input and output in the other works . Moreover, finite‐horizon H‐infinity state estimation was investigated for periodic neural networks over fading channels in the work of Li et al…”

Section: Introductionmentioning

confidence: 99%

Adaptive neural output feedback control for uncertain nonlinear systems with input quantization and output constraints

Wang

Zhang

Yang

2020

Adaptive Control & Signal

View full text Add to dashboard Cite

Summary In this paper, we are concerned with the problem of adaptive output‐feedback tracking control for nonlinear systems with input quantization, unmodeled dynamics, and output constraints. A novel quantizer with the advantages of hysteresis and uniform quantizer is introduced to handle input signals. A barrier Lyapunov function is employed to solve the output constraints. The state unmodeled dynamics is solved by using a Lyapunov description, and neural networks are used to approximate the unknown smooth functions produced in the adaptive control design process. The controller design is simplified by combining the new quantizer with dynamic surface control method. The mathematical derivation shows the stability of the closed‐loop system and the effectiveness of output constraints. Simulation illustrates and clarifies the theoretical findings.

show abstract

“…Adaptive control can only be applied to some special systems, that is, parametric uncertainties satisfying linear growth conditions . Alternative robust control methods, such as H ∞ control, have strong robustness in dealing with large‐scale operating regions even in the presence of disturbance scenarios, but the design process is complex and the synthesis can be difficult and time‐consuming …”

Section: Introductionmentioning

confidence: 99%

Decentralized sliding mode control for a class of nonlinear interconnected systems by static state feedback

Feng

Zhao

Yan

et al. 2020

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

Summary In this paper, a class of interconnected systems is considered, where the nominal isolated systems are fully nonlinear. A robust decentralized sliding mode control based on static state feedback is developed. By local coordinate transformation and feedback linearization, the interconnected system is transformed to a new regular form. A composite sliding surface which is a function of the system state variables is proposed and the stability of the corresponding sliding mode dynamics is analyzed. A new reachability condition is proposed and a robust decentralized sliding mode control is then designed to drive the system states to the sliding surface in finite time and maintain a sliding motion thereafter. Both uncertainties and interconnections are allowed to be unmatched and are assumed to be bounded by nonlinear functions. The bounds on the uncertainties and interconnections have more general forms when compared with existing work. A MATLAB simulation example is used to demonstrate the effectiveness of the proposed method.

show abstract

Finite-Horizon $H_{\infty}$ State Estimation for Periodic Neural Networks Over Fading Channels

Cited by 77 publications

References 46 publications

Multigradient recursive reinforcement learning NN control for affine nonlinear systems with unmodeled dynamics

Multigradient recursive reinforcement learning NN control for affine nonlinear systems with unmodeled dynamics

Adaptive neural output feedback control for uncertain nonlinear systems with input quantization and output constraints

Decentralized sliding mode control for a class of nonlinear interconnected systems by static state feedback

Contact Info

Product

Resources

About