Identification and decentralized adaptive control using dynamical neural networks with application to robotic manipulators

Karakasoglu, A.; Sudharsanan, Subramania I.; Sundareshan, Malur K.

doi:10.1109/72.286887

Cited by 108 publications

(45 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, for the higher order subgrids, use the smaller radius, i.e., (16) Usually, choose (17) where is a constant and less than one. Thus, the chosen centers from the set are given by the set and (18) In order that the basis function candidates in the set that are less than an activation threshold to the nearest grid node in the th subgrid are outside the set , it can be deduced from (8) and (15) that the must be chosen to be (19) for , where represents the activation threshold.…”

Section: Selection Of Basis Functionsmentioning

confidence: 99%

“…A large number of control structures have been proposed, including supervised control [55], direct inverse control [34], model reference control [39], internal model control [13], predictive control [14], [56], [29], gain scheduling [12], optimal decision control [10], adaptive linear control [7], reinforcement learning control [1], [3], variable structure control [30], indirect adaptive control [39], and direct adaptive control [19], [45], [50], [51]. The principal types of neural networks used for control problems are the multilayer perceptron (MLP) neural networks with sigmoidal units [34], [39], [48] and the radial basis function (RBF) neural networks [41], [43], [47].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Variable neural networks for adaptive control of nonlinear systems

Liu

Kadirkamanathan

Billings

1999

IEEE Trans. Syst., Man, Cybern. C

View full text Add to dashboard Cite

Abstract-This paper is concerned with the adaptive control of continuous-time nonlinear dynamical systems using neural networks. A novel neural network architecture, referred to as a variable neural network, is proposed and shown to be useful in approximating the unknown nonlinearities of dynamical systems. In the variable neural networks, the number of basis functions can be either increased or decreased with time according to specified design strategies so that the network will not overfit or underfit the data set. Based on the Gaussian radial basis function (GRBF) variable neural network, an adaptive control scheme is presented. The location of the centers and the determination of the widths of the GRBF's in the variable neural network are analyzed to make a compromise between orthogonality and smoothness. The weight adaptive laws developed using the Lyapunov synthesis approach guarantee the stability of the overall control scheme, even in the presence of modeling error. The tracking errors converge to the required accuracy through the adaptive control algorithm derived by combining the variable neural network and Lyapunov synthesis techniques. The operation of an adaptive control scheme using the variable neural network is demonstrated using two simulated examples.Index Terms-Adaptive control, neural networks, nonlinear systems, radial basis functions.

show abstract

Section: Selection Of Basis Functionsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Variable neural networks for adaptive control of nonlinear systems

Liu

Kadirkamanathan

Billings

1999

IEEE Trans. Syst., Man, Cybern. C

View full text Add to dashboard Cite

show abstract

“…Fuzzy techniques have been recently used to effectively approximate unknown nonlinear dynamics [9]- [11]. However, in the conventional adaptive fuzzy control, the desired tracking performance cannot be guaranteed from the more theoretical control perspective.…”

mentioning

confidence: 99%

Robust tracking enhancement of robot systems including motor dynamics: a fuzzy-based dynamic game approach

Chen

Uang

Tseng

1998

IEEE Trans. Fuzzy Syst.

115

View full text Add to dashboard Cite

Abstract-A robust tracking control design of robot systems including motor dynamics with parameter perturbation and external disturbance is proposed in this study via adaptive fuzzy cancellation technique. A minimax controller equipped with a fuzzy-based scheme is used to enhance the tracking performance in spite of system uncertainties and external disturbance. The design procedure is divided into three steps. At first, a linear nominal robotic control design is obtained via model reference tracking with desired eigenvalue assignment. Next, a fuzzy logic system is constructed and then tuned to eliminate the nonlinear uncertainties as possibly as it can to enhance the tracking robustness. Finally, a minimax control scheme is specified to optimally attenuate the worst-case effect of both the residue due to fuzzy cancellation and external disturbance to achieve a minimax tracking performance. In addition, an adaptive fuzzybased dynamic game theory is introduced to solve the minimax tracking problem. The proposed method is appropriate for the robust tracking design of robotic systems with large parameter perturbation and external disturbance. A simulation example of a two-link robotic manipulator driven by dc motors is also given to demonstrate the effectiveness of proposed design method's tracking performance.Index Terms-Fuzzy-based dynamic game theory, fuzzy cancellation, tracking enhancement, uncertain robot systems.

show abstract

“…Neural net techniques have been successfully applied in various fields such as function approximation, signal processing and adaptive (or) learning control for nonlinear systems. Using neural networks, a variety of off-line learning control algorithms have been developed for nonlinear systems [17,25]. A variety of numerical algorithms have been developed for solving the algebraic Riccati equation.…”

Section: Introductionmentioning

confidence: 99%

Solution of matrix Riccati differential equation for nonlinear singular system using neural networks

Samath¹,

Selvaraju²

2010

IJCA

View full text Add to dashboard Cite

In this paper, the solution of the matrix Riccati differential equation(MRDE) for nonlinear singular system is obtained using neural networks. The goal is to provide optimal control with reduced calculus effort by comparing the solutions of the MRDE obtained from well known traditional Runge Kutta(RK)method and nontraditional neural network method. Accuracy of the neural solution to the problem is qualitatively better. The advantage of the proposed approach is that, once the network is trained, it allows instantaneous evaluation of solution at any desired number of points spending negligible computing time and memory. The computation time of the proposed method is shorter than the traditional RK method. An illustrative numerical example is presented for the proposed method.

show abstract

Identification and decentralized adaptive control using dynamical neural networks with application to robotic manipulators

Cited by 108 publications

References 20 publications

Variable neural networks for adaptive control of nonlinear systems

Variable neural networks for adaptive control of nonlinear systems

Robust tracking enhancement of robot systems including motor dynamics: a fuzzy-based dynamic game approach

Solution of matrix Riccati differential equation for nonlinear singular system using neural networks

Contact Info

Product

Resources

About