arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

R. Saeks

45 papers · 1969–2016

1969: 21970: 21971: 01972: 01973: 11974: 21975: 11976: 31977: 61978: 21979: 21980: 31981: 01982: 21983: 01984: 11985: 11986: 01987: 01988: 11989: 01990: 01991: 01992: 01993: 01994: 11995: 11996: 01997: 01998: 11999: 12000: 02001: 12002: 72003: 22004: 02005: 02006: 02007: 02008: 02009: 02010: 02011: 02012: 02013: 02014: 12015: 02016: 1
1969peak 7 in 20022016

Frequent co-authors

Papers

showing 1–25 of 45 RSS

Adaptive dynamic programming

J. Murray, C. Cox, G.G. Lendaris +1 more

Computational Theory and Mathematics · IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews) · · cited 675× · DOI ↗ · more like this ·

Unlike the many soft computing applications where it suffices to achieve a "good approximation most of the time," a control system must be stable all of the time. As such, if one desires to learn a control law in real-time, a fusion of soft computing techniques to learn the appropriate control law with hard computing techniques to maintain the stability constraint and guarantee convergence is required. The objective of the paper is to describe an adaptive dynamic programming algorithm (ADPA) which fuses soft computing techniques to learn the optimal cost (or return) functional for a stabilizable nonlinear system with unknown dynamics and hard computing techniques to verify the stability and convergence of the algorithm. Specifically, the algorithm is initialized with a (stabilizing) cost functional and the system is run with the corresponding control law (defined by the…

Feedback system design: The fractional representation approach to analysis and synthesis

C. Desoer, Ruey‐Wen Liu, J. Murray +1 more

Computational Theory and Mathematics · IEEE Transactions on Automatic Control · · cited 650× · DOI ↗ · more like this ·

The problem of designing a feedback system with prescribed properties is attacked via a fractional representation approach to feedback system analysis and synthesis. To this end we let <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</tex> denote a ring of operators with the prescribed properties and model a given plant as the ratio of two operators in <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</tex> . This, in turn, leads to a simplified test to determine whether or not a feedback system in which that plant is embedded has the prescribed properties and a complete characterization of those compensators which will "place" the feedback system in <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</tex> . The theory is formulated axiomatically…

Fractional representation, algebraic geometry, and the simultaneous stabilization problem

R. Saeks, J. Murray

Geometry and Topology · IEEE Transactions on Automatic Control · · cited 277× · DOI ↗ · more like this ·

An explicit relationship between the fractional representation approach feedback system design and the algebro-geometric approach to system theory, is formulated and used to derive a global solution to the feedback system problem. These techniques are then applied to the simultaneous stabilization problem, yielding a natural geometric criterion for a set of plants to be simultaneously stabilized by a single compensator.

Multivariable Nyquist theory

Raymond A. DeCarlo, J. Murray, R. Saeks

Signal Processing · International Journal of Control · · cited 110× · DOI ↗ · more like this ·

It is shown that a multivariable digital filter is stable if and only if A(z1, z2…,zn)≠0 when | z1 1| = | z2|=… = |zn| = 1 and the Nyquist plot for the single-variable function A(z, z, … z) does not encircle zero. Here A(z1, z2… zn) is the denominator polynomial of the relatively prime rational transfer function of the digital filter.

Causality in Hilbert Space

R. Saeks

Mathematical Physics · SIAM Review · · cited 66× · DOI ↗ · more like this ·

A new formalism, termed the resolution space, is presented within which the theory of causal systems may be unified and extended. The resulting formalism, which is defined as a Hilbert space together with a resolution of the identity, readily includes the commonly encountered function and sequence space causality concepts yet is sufficiently straightforward to allow the various aspects of network and system theory which are dependent on the time parameter to be studied in an operator theoretic context without the detailed structure of a function space. Specific results include additive and multiplicative decomposition theorems for causal operators which naturally extend the “realizable part” and “spectral” decompositions of classical system theory and an integral representation theorem for linear operators on a resolution space. The general theory is illustrated with a number of…

Analog Fault Diagnosis with Failure Bounds

Chiwan-Chia Wu, K. Nakajima, Chin‐Long Wey +1 more

Hardware and Architecture · IEEE Transactions on Circuits and Systems · · cited 58× · DOI ↗ · more like this ·

A simulation-after-test algorithm for the analog fault diagnosis problem is proposed in which a bound on the maximum number of simultaneous failures is used to minimize the number of test points required. The resultant algorithm is applicable to both linear and nonlinear systems with multiple hard or soft faults and can be used to isolate failures up to an arbitrarily specified "replaceable chip or subsystem."

Linear Hopfield networks and constrained optimization

G.G. Lendaris, Karl Mathia, R. Saeks

Artificial Intelligence · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · · cited 49× · DOI ↗ · more like this ·

It is shown that a Hopfield neural network (with linear transfer functions) augmented by an additional feedforward layer can be used to compute the Moore-Penrose generalized inverse of a matrix. The resultant augmented linear Hopfield network can be used to solve an arbitrary set of linear equations or, alternatively, to solve a constrained least squares optimization problem. Applications in signal processing and robotics are considered. In the former case the augmented linear Hopfield network is used to estimate the "structured noise" component of a signal and adjust the parameters of an appropriate filter on-line, whereas in the latter case it is used to implement an on-line solution to the inverse kinematics problem.

A search algorithm for the solution of the multifrequency fault diagnosis equations

H. Chen, R. Saeks

Hardware and Architecture · IEEE Transactions on Circuits and Systems · · cited 40× · DOI ↗ · more like this ·

A search algorithm for the solution of the fault diagnosis equations arising in linear time invariant analog circuits and systems is presented. By exploitation of Householder's formula an efficient algorithm whose computational complexity is a function of the number of system failures rather than the number of system components is obtained.

On the Implementation of an Analog ATPG: The Linear Case

Chin‐Long Wey, R. Saeks

Hardware and Architecture · IEEE Transactions on Instrumentation and Measurement · · cited 38× · DOI ↗ · more like this ·

A self-testing algorithm in which post-test simulation with failure bounds is employed, has been proposed. Based on this self-testing algorithm, an analog Automatic Test Program Generation (ATPG) for linear circuits or systems is being developed. The AATPG code is subdivided into off-line and on-line components while the actual test can be run in either a fully automatic mode or interactively.

The factorization problem—A survey

R. Saeks

Mathematical Physics · Proceedings of the IEEE · · cited 37× · DOI ↗ · more like this ·

The problem of factoring a positive Hermitian operator into the product of a causal operator and its adjoint is reviewed. Three classes of factorization are studied, miniphase factorization, regular factorization, and special factorzation. The former always exists, and a complete representation thereof is obtained in terms of the properties of reproducing kernel spaces. The regular factorization does not exist, in general, and its existence theroy is shown to be equivalent to the unitary equivalence problem of classical operator theory. Finally, an existence and uniqueness theorem for special factors is formulated in terms of certain Cauchy integrals and is shown to be closely related to Wiener-Hopf theory.

Continuation methods in circuit analysis

Kwong-Shu Chao, R. Saeks

Computational Theory and Mathematics · Proceedings of the IEEE · · cited 33× · DOI ↗ · more like this ·

The use of continuation methods in the computer-aided analysis of electronic circuits is surveyed. Such methods are especially suitable when one desires to compute the solutions to a family of circuit analysis problems as a function of a continuous parameter. Applications of the concept to the location of multiple solutions to nonlinear equations, the computation of input-output characteristics for nonlinear networks, large-change sensitivity analysis, and the computation of multivariable Nyquist plots are reviewed.

On the implementation of an analog ATPG: the nonlinear case

C.-L. Wey, R. Saeks

Hardware and Architecture · IEEE Transactions on Instrumentation and Measurement · · cited 24× · DOI ↗ · more like this ·

In order to reduce the complexity of the fault diagnosis equations and still retain computational simplicity, a self-testing algorithm has been proposed and implemented on a VMS VAX 11/780 for linear circuits. A prototype implementation of such an algorithm for nonlinear circuits and systems is presented. The proposed analog automatic test program generator (AATPG) for nonlinear circuits and systems is divided into offline and online processes. Unlike the simulation of the pseudocircuits in the linear case, which can be achieved by a matrix/vector multiplication, the circuit simulator SPICE is used to simulate the nonlinear pseudocircuits. The automatic SPICE code generator required for this simulation is presented. The proposed AATPG for nonlinear circuits has been implemented on a VMS VAX 11/780. The actual test can be run in either a fully automatic mode or interactively.< <ETX…

Globally convergent approximate dynamic programming applied to an autolander

J. Murray, C. Cox, R. Saeks +1 more

Computational Theory and Mathematics · PDXScholar (Portland State University) · · cited 21× · DOI ↗ · more like this ·

A globally convergent nonlinear Approximate Dynamic Programming algorithm is described, and an implementation of the algorithm in the linear case is developed. The resultant linear Approximate Dynamic Programming algorithm is illustrated via the design of an autolander for the NASA X-43 research aircraft, without a priori knowledge of the X-43's flight dynamics.

The encirclement condition An approach using algebraic topology†

Raymond A. DeCarlo, R. Saeks

Geometry and Topology · International Journal of Control · · cited 18× · DOI ↗ · more like this ·

Abstract The usual proof of the Nyquist theorem depends heavily on the argument principle. The argument principle, however, supplies unneeded information in that it counts the number of encirclements of ‘ − 1 ’. System stability requires only a binary decision on the number of encirclements. Using homotopy theory, a branch of algebraic topology, we construct new proofs of the classical Nyquist criteria along with the more recent linear multivariable results. In both cases the proof is essentially the same. Moreover, we believe our point of view to be more intuitive and capable of generalization to systems characterized by functions of several complex variables.

The Adaptive Dynamic Programming Theorem

J. Murray, C. Cox, R. Saeks

Computational Theory and Mathematics · Birkhäuser Boston eBooks · · cited 17× · DOI ↗ · more like this ·

The centerpiece of the theory of dynamic programming is the HamiltonJacobi-Bellman (HJB) equation, which can be used to solve for the optimal cost functional Vo for a nonlinear optimal control problem, while one can solve a second partial differential equation for the corresponding optimal control law ko.Although the direct solution of the HJB equation is computationally untenable, the HJB equation and the relationship between Vo and koserves as the basis for the adaptive dynamic programming algorithm. Here, one starts with an initial cost functional and stabilizing control law pair (Vo, k0) and constructs a sequence of cost functional/control law pairs (Vi, ki) in real time, which are stepwise stable and converge to the optimal cost functional/control law pair, for a prescribed nonlinear optimal control problem with unknown input affine state dynamics.

Asymptotic dynamic programming: preliminary concepts and results

R. Saeks, C. Cox, Karl Mathia +1 more

Computational Theory and Mathematics · Proceedings of International Conference on Neural Networks (ICNN'97) · · cited 17× · DOI ↗ · more like this ·

A formal setting for the development of adaptive critic techniques is established in a nonlinear dynamical systems setting and some preliminary stability and suboptimality theorems are developed. Two alternative versions of the theory are developed, one with a known plant and one with an initial unknown plant which must be identified online.

Solving Nonlinear Equations Using Recurrent Neural Networks

Karl Mathia, R. Saeks

Artificial Intelligence · · cited 16× · OpenAlex ↗ · more like this ·

A class of recurrent neural networks is developed to solve nonlinear equations, which are approximated by a multilayer perceptron (MLP). The recurrent network includes a linear Hopfield network (LHN) and the MLP as building blocks. This network inverts the original MLP using constrained linear optimization and Newton’s method for nonlinear systems. The solution of a nonlinear equation with computer simulation illustrates the algorithm. 1 Problem Context A class of recurrent neural networks is used to solve nonlinear equations, where the motivation for using artificial neural networks is their learning capability. For this work the properties of multilayer perceptron (MLP) as universal function approximators [3] are of particular advantage: data with unknown structure, taken from a nonlinear system, are approximated by an MLP. The MLP is then inverted, i.e. the approximated equation is…

Adaptive critic control and functional link neural networks

C. Cox, R. Saeks

Computational Theory and Mathematics · · cited 14× · DOI ↗ · more like this ·

An adaptive critic neural control algorithm, employing an analytic relationship between the optimal cost function and the optimal control law is defined. This approach reduces the complexity of the adaptive critic algorithm, by replacing the iteration on the control law with the solution of a nonlinear equation, while simultaneously facilitating the development of stability and robustness criteria. The resultant algorithm can be implemented without a priori knowledge of the plant dynamics, yielding an adaptive control algorithm which does not require an explicit system identification process. Finally, the algorithm is specialized to the linear case where it is realized by a system of three coupled functional link neural networks.

A Nyquist-like test for the stability of two-dimensional digital filters

Raymond A. DeCarlo, R. Saeks, J. Murray

Signal Processing · Proceedings of the IEEE · · cited 14× · DOI ↗ · more like this ·

This paper constructs a Nyquist-like test for the stability of two-dimensional digital filters. The test takes the form of a continuum of classical one-variable Nyquist plots parameterized by the elements of the unit circle of the complex plane. Since the parameter space is compact, the test can be accurately approximated by a finite number of classical one-variable Nyquist plots and is therefore readily implemented on a computer.

A radial basis function implementation of the adaptive dynamic programming algorithm

G. Lendaris, C. Cox, R. Saeks +1 more

Computational Theory and Mathematics · PDXScholar (Portland State University) · · cited 11× · DOI ↗ · more like this ·

Adaptive Dynamic Programming (ADP) constitutes a potentially powerful approach to optimal control. An approximation to the Bellman cost functional is updated in real time. The technique is applicable to a broad class of nonlinear networks with unknown dynamics and is guaranteed to converge to the optimal control with stepwise stability. The goal of this paper is to describe an implementation of the ADP algorithm in which a radial basis function is used to define the approximate cost functional, which is updated locally in the neighborhood of the state trajectory each time the system is run. An application of the algorithm to a nonlinear flight control problem with unknown aircraft dynamics is presented.

Neural networks and fault diagnosis in rotating machinery

Kevin L. Priddy, M. Lothers, R. Saeks

Artificial Intelligence · · cited 9× · DOI ↗ · more like this ·

A series of experiments in which neural networks are used to detect and classify faults in rotating machinery are described. Specific experimental testbeds included a helicopter tail rotor gearbox, fire pumps and condensation pumps.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

On the Structure of Invertible Operators in a Nest-Subalgebra of a von Neumann Algebra

Gareth J. Knowles, R. Saeks

Applied Mathematics · Birkhäuser Basel eBooks · · cited 7× · DOI ↗ · more like this ·

The topological structure of the collection of all invertible elements of certain operator algebras is investigated. The first such algebra A will be a nest subalgebra of a von Neumann algebra. It is shown that, the collection of invertible elements of A with inverse also in A, satisfying certain boundary conditions, are in the principal component of the identity when assigned the strong operator topology. The second /A is a subalgebra of A. When assigned the uniform topology, it is shown that the invertible elements of /A are in the principal component of the identity. The results are applied to a large variety of examples where /A is shown to be extensive.

Fourier Analysis in Hilbert Spaces

R. Saeks

Applied Mathematics · SIAM Review · · cited 7× · DOI ↗ · more like this ·

The techniques of non-self adjoint spectral theory are employed to formulate a frequency domain theory for general linear systems characterized by operators on a (Hilbert) resolution space. The properties and relationships between the various frequency domain representations are studied and the special forms which these representations take when applied to causal and passive operators are delineated. Finally, the theory is illustrated via the formulation of a number of operator decomposition problems in a frequency domain setting and a study of the frequency domain stability problem.

Continuations approach to large-change sensitivity analysis

R. Saeks, K.S. Chao

Numerical Analysis · IEE Journal on Electronic Circuits and Systems · · cited 6× · DOI ↗ · more like this ·

A differential equation whose trajectories represent the evolution of the transfer function of an electronic circuit or system as a function of some potentially variable parameter is derived and used as the basis of a continuations algorithm for large-change sensitivity analysis. The resultant algorithm is computationally efficient and is illustrated with several examples.

An Approach to Built-In Testing

R. Saeks

Hardware and Architecture · IEEE Transactions on Aerospace and Electronic Systems · · cited 5× · DOI ↗ · more like this ·

The archiitecture and justification for an approach to built-in testing (BIT) in electronic circuits and systems is presented. The proposed system is capable of on-line fault detection and prediction up to the shop replaceable assembly (SRA) level and is designed to interface with external automatic test equipment (ATE) for off-line fault diagnosis within the SRA. The constituent parts of the BIT system have been extensively simulated and the approach appears to be suitable for hardware implementation both with respect to computational and economic considerations.