A sliding mode strategy for adaptive learning in Adalines

Sira-Ramírez, H.; Colina-Morles, Eliezer

doi:10.1109/81.481195

Cited by 68 publications

(14 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The remaining three subplots in Fig. 4 depict the time evolution of the adjustable neurocontroller parameters These subplots confirm the evolution in finite volume claim of [6,9]. The variables seen in Figs.…”

Section: Simulation Resultssupporting

confidence: 63%

“…The problem of parameter tuning in RBFNN has extensively been studied in the literature. Some important ones of which are the gradient descent technique (Error Backpropagation) [4], Levenberg-Marquardt algorithm [5], and hybrid methods such as Variable Structure Systems (VSS) theory based learning strategies [6][7][8]. At a first glance, what a reader notices is the fact that the application of above mentioned approaches for tuning the parameters of a controller require the target value of the control signal, which is unavailable by the nature of the problem.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Radial basis function neural networks in variable structure control of a class of biochemical processes

Efe

Kaynak

Wilamowski³

et al.

IECON'01. 27th Annual Conference of the IEEE Industrial Electronics Society (Cat. No.37243)

View full text Add to dashboard Cite

Biochemical processes often display a complicated dynamic behavior, the detailed understanding of which frequently constitutes a barrier between the theoretical foundations and practical implementations. One way of handling the complexity is to use intelligent approaches in the design of controllers. This paper presents an analytic approach to design controllers based on Radial Basis Function Neural Networks (RBFNN) with particular emphasis on the extraction of the error measure to be used in parameter tuning. The simulation studies stipulate that the control system exhibits a highly robust behavior against disturbances and sharp changes in the command signal. The most important contribution of the paper is that the method presented does not require the analytical details describing the plant dynamics available.

show abstract

Section: Simulation Resultssupporting

confidence: 63%

Section: Introductionmentioning

confidence: 99%

Radial basis function neural networks in variable structure control of a class of biochemical processes

Efe

Kaynak

Wilamowski³

et al.

IECON'01. 27th Annual Conference of the IEEE Industrial Electronics Society (Cat. No.37243)

View full text Add to dashboard Cite

show abstract

“…The use of an NN for the calculation of the equivalent control term is also proposed by Jezernik, Rodic, Safaric, and Curket [27]. Ramirez and Morles [28] propose a dynamical sliding mode control approach for robust adaptive learning in analog adaptive linear elements (Adalines). Some other reports dealing with the integration of NN and SMC have appeared in the literature [29][30][31][32][33].…”

Section: Fnnsmc Designmentioning

confidence: 98%

Fuzzy neural network sliding mode control for long delay time systems based on fuzzy prediction

2007

Neural Comput Applic

View full text Add to dashboard Cite

Delay time, which may degrade the control performance, is frequently encountered in various control processes. The fuzzy neural network sliding mode controller (FNNSMC), which incorporates the fuzzy neural network (FNN) with the sliding mode controller (SMC), is developed to control the long delay system with unknown model based on fuzzy prediction algorithm in the paper. According to the characteristics of the long delay systems, we simulate the manual operating process and predict the delayed error and its derivative based on the information of the input and output variables of the process, and then feedback these prediction values to the FNN and train the FNN with the regulation function by the idea of sliding mode control until the better control results are obtained. The FNNSMC has more robustness due to the abilities of the learning and reasoning and can eliminate the drawbacks of the general SMC, namely the chattering in the control signal and the needing knowledge of the bounds of the disturbances and uncertainties. Simulation examples demonstrate the advantages of the proposed control scheme.

show abstract

“…The method presented in Ref. 33 also presents the forward and inverse dynamics identification of a Kapitsa pendulum. The fundamental difference of the algorithm presented in this paper is the fact that the derivation is based on the mixture of two different update values.…”

Section: Introductionmentioning

confidence: 99%

Stabilizing and robustifying the learning mechanisms of artificial neural networks in control engineering applications

Efe¹,

Kaynak²

2000

Int. J. Intell. Syst.

View full text Add to dashboard Cite

This paper discusses the stabilizability of artificial neural networks trained by utilizing the gradient information. The method proposed constructs a dynamic model of the conventional update mechanism and derives the stabilizing values of the learning rate. The stability in this context corresponds to the convergence in adjustable parameters of the neural network structure. It is shown that the selection of the learning rate as imposed by the proposed algorithm results in stable training in the sense of Lyapunov. Furthermore, the algorithm devised filters out the high frequency dynamics of gradient descent method. The excitation of this dynamics typically occurs in the presence of noise and abruptly changing parameters of the mapping being learned. This adversely influences the learning performance that can be attained during a training cycle. A natural consequence following this excitation is divergence in parameter space. The method analyzed in this paper integrates the gradient descent technique with variable structure systems methodology, which is well known with its robustness to environmental disturbances. In the simulations, control of a three degrees of freedom anthropoid robot is chosen for the evaluation of the performance. For this purpose, a feedforward neural network structure is utilized as the controller. Highly nonlinear dynamics of the plant, existence of a considerable amount of observation noise and the adverse effects of gravitational forces constitute the difficulties to be alleviated by the neurocontroller trained with the proposed method. In order to come up with a fair comparison, the results obtained with the pure gradient descent technique with the same initial conditions are also presented and discussed.

show abstract

A sliding mode strategy for adaptive learning in Adalines

Cited by 68 publications

References 7 publications

Radial basis function neural networks in variable structure control of a class of biochemical processes

Radial basis function neural networks in variable structure control of a class of biochemical processes

Fuzzy neural network sliding mode control for long delay time systems based on fuzzy prediction

Stabilizing and robustifying the learning mechanisms of artificial neural networks in control engineering applications

Contact Info

Product

Resources

About