Showing papers on "System identification published in 1970"
Book•
01 Jan 1970
TL;DR: The success can be started by knowing the basic knowledge and do actions, and the system identification for self adaptive control will really give you the good idea to be successful.
Abstract: By reading, you can know the knowledge and things more, not only about what you get from people to people. Book will be more trusted. As this system identification for self adaptive control, it will really give you the good idea to be successful. It is not only for you to be success in certain life you can be successful in everything. The success can be started by knowing the basic knowledge and do actions.
226 citations
TL;DR: A survey of several techniques for the identification of dynamic systems using computer techniques, including spectral analysis, search and gradient methods, quasi-linearization, and stochas tic approximation, particularly suited to digital or hybrid computer implementation.
Abstract: This paper presents a survey of a number of techniques for the identification of dynamic systems using computer techniques. The techniques discussed are particularly suited to digital or hybrid com...
97 citations
TL;DR: In this paper, the authors examined the asymptotic properties of the unknown parameters and the unknown initial state of linear, stable, constant coefficient, discrete time dynamic systems where plant noise and observation noise are present.
Abstract: This paper examines the asymptotic properties of the maximum likelihood estimates of the unknown parameters and the unknown initial state of linear, stable, constant coefficient, discrete time dynamic systems where plant noise and observation noise are present. Necessary and sufficient conditions are obtained for the system parameter estimates to converge with probability one, to be asymptotically normal and to converge in mean square. These conditions require that the system representation be unique and impose a simple constraint on the input sequence. Under these conditions, the initial state estimate is shown to be asymptotically unbiased and have finite covariance.
53 citations
01 Jan 1970
TL;DR: System identification, considering input signals classification, model structure, linear/nonlinear systems identifying and on-line/real-time techniques as mentioned in this paper, considering input signal classification and model structure.
Abstract: System identification, considering input signals classification, model structure, linear/nonlinear systems identifying and on-line/real time techniques
46 citations
TL;DR: In this article, an automatic method is presented for the identification of time delays and parameters in linear systems, where unknown parameters in these equations are determined by a continuous steepest descent minimization of an error function.
Abstract: An automatic method is presented for the identification of time delays and parameters in linear systems. It is assumed that processes to be identified can be described by linear differential-difference equations. Unknown parameters in these equations are determined by a continuous steepest descent minimization of an error function. If the process parameters are constant, a region exists in the parameter space in which the tracking paramters converge to the process parameters. If all process time delays are known as well, the tracking parameters will always converge. Two examples are presented which show that the method may be used to identify slowly varying parameters. A third example indicates techniques leading to a more economical implementation of the method.
37 citations
TL;DR: In this article, an identification procedure for determination of coefficients of pulse transfer function is proposed, which is applicable for the case when measurement noises of input and output are independent of input to a process and the variances of the noises are known only in the form of the ratio.
Abstract: An identification procedure for determination of coefficients of pulse transfer function is proposed. This procedure is applicable for the case when measurement noises of input and output are independent of input to a process and the variances of the noises are known only in the form of the ratio. The noises can be colored. The criterion used for identification is given by extending the mean-square equation error. The optimal estimates are obtained from the eigenvector of a linear operator which corresponds to the least eigenvalue. The theoretical analysis is given and the direction for the practical usage is suggested to use the least-square error criterion.
14 citations
TL;DR: A unified exposition of certain computer-aided statistical estimators for control systems identification is presented, together with some numerical results which show and compare the effectiveness of these estimators.
Abstract: A unified exposition of certain computer-aided statistical estimators for control systems identification is presented, together with some numerical results which show and compare the effectiveness of them
8 citations
TL;DR: A stochastic approximation algorithm is presented for the identification of the pointwise approximation of Green's function in the presence of additive measurement noise.
Abstract: The problem of identifying Green's function of a linear time-invariant distributed parameter system is considered. A discrete lumped version of the distributed system is developed for which a pointwise approximation of Green's function is the state transition matrix. A stochastic approximation algorithm is presented for the identification of the pointwise approximation of Green's function in the presence of additive measurement noise.
7 citations
TL;DR: In this article, a new method is proposed that can identify a linear system with a random input signal in a short time for application to the identification of manual tracking systems whose response characteristics vary slowly with time.
Abstract: A new method is proposed that can identify a linear system with a random input signal in a short time for application to the identification of manual tracking systems whose response characteristics vary slowly with time. Making use of this method of identification, a display system including an on-line computer is constructed, which displays the slowly varying response characteristics of manual tracking systems on a CRT screen in real time. A software system that facilitates performance of various kinds of tracking experiments is developed. Finally, some results of simple pursuit tracking experiments are described and the mode of the tracking operation is discussed.
6 citations
TL;DR: In this article, conditions are established under which a problem of simultaneous identification and control may be treated by solving two separate problems: one an estimation problem on the set of unknown parameters of the plant as well as its external disturbances, and the other a control problem.
Abstract: Conditions are established under which a problem of simultaneous identification and control may be treated by solving two separate problems: one an estimation problem on the set of unknown parameters of the plant as well as its external disturbances, and the other a control problem. This property is referred to as a separation and, when it is applicable, solutions can be readily obtained. An example is presented to illustrate how the property may be applied to specific problems.
4 citations
TL;DR: In this paper, a general method for calculating the parameters of a 4-level aperiodic input used in a scheme proposed recently for identification of the linear portions of a system containing a zero-memory nonlinearity was described.
Abstract: A general method is described for calculating the parameters of a 4-level aperiodic input used in a scheme proposed recently for identification of the linear portions of a system containing a zero-memory nonlinearity.
01 Apr 1970
TL;DR: In this article, the feasibility of applying a Newtonian system identification technique to a nonlinear three degree of freedom system of equations describing the steering and maneuvering of a surface ship is investigated.
Abstract: : The feasibility of applying a Newtonian system identification technique to a nonlinear three degree of freedom system of equations describing the steering and maneuvering of a surface ship is investigated. The input to the system identification program is provided by both analog and digital computer generated maneuvers. The dominant eleven coefficients, treated as unknowns, are sought using exact and noisy data, and the utility and limitations of the method are described. (Author)
TL;DR: In this article, a nonparametric stochastic approximation type of algorithm for on-line identification is considered, which is useful when the degree of inaccuracy in the knowledge of the system is much larger than possible output measurement errors.
Abstract: A nonparametric stochastic approximation type of algorithm for on-line identification is considered. By on-line, it is meant that the system is driven by an arbitrary, known input, the control, in addition to an input noise disturbance. Thus, the control effort proceeds in conjunction with identification and is in no way restricted by the requirements of identification. The system is assumed to be all-pole with unknown poles, unknown gain, and noiseless output measurement. Consequently, the procedure is useful when the degree of inaccuracy in the knowledge of the system is much larger than possible output measurement errors. Adaptive control using this procedure is discussed, together with a numerical example.
TL;DR: A control algorithm which incorporates the identification of a concurrent real-time system identification with a computationally simpler forward-calculation correction technique.
Abstract: A recent publication [1] proposes a solution requiring complete recalculation of a sequence of matrices (backward from the end of the interval) with each model change provided by the concurrent real-time system identification. The new result here is a control algorithm which incorporates the identification with a computationally simpler forward-calculation correction technique.
Abstract: Recursive algorithms for online identification of discrete-time systems have been described. These provide minimum-norm estimates of the parameter vector when insufficient data are available, and least-squares estimates with sufficient data. Matrix inversion is not required; nor is the a priori knowledge of noise statistics or probability densities of the parameters.
TL;DR: In this paper, a technique for correcting the feedback control coefficients with each change (in the mean of the model) provided by concurrent system identification is presented. But this technique does not consider the covariance of the system model.
Abstract: The preceding correspondence [1] gives a technique for correcting the feedback control coefficients with each change (in the mean of the model) provided by concurrent system identification. The new result here is a similar technique that incorporates changes in both the mean and covariance of the system model.
TL;DR: In this article, a new substructure system identification algorithm, which produces a linear, viscous-damped, reduced-order physical model (i.e., A/, C, and /f), is described and the results of numerical simulations used to test the proposed new algorithm are presented.
Abstract: When a complex structural system must be analyzed for its response to dynamic excitation, some form of substructure coupling method, or component mode synthesis (CMS) method, is usually employed. It is generally necessary to perform some form of vibration test to validate the separate substructure math models, which are then coupled together for the system analysis. When such tests are performed, it is important to give special attention to the way that the substructure is supported and the way that it is excited. A new substructure system identification algorithm, which produces a linear, viscous-damped, reduced-order physical model (i.e., A/, C, and /f), is described in this paper, and the results of numerical simulations used to test the proposed new algorithm are presented. Of particular interest is the comparison between the results obtained by using the ordinary least-squares (OLS) method and those based on the total least-squares (TLS) method.
TL;DR: In this paper, a finite-dimensional solution for a system identification problem is derived using Bucy's representation theorem for conditional density functions, showing how, for this nonlinear estimation problem, the function-space representation for the posterior density can be greatly simplified.
Abstract: Using Bucy's representation theorem for conditional density functions, a finite-dimensional solution for a system identification problem is derived. The derivation displays how, for this nonlinear estimation problem, the function-space representation for the posterior density can be greatly simplified.
TL;DR: This paper presents a stochastic gradient procedure, a method of estimating the minimum of a criterion function despite the random effects of noise, which has potential application to learning systems, parameter estimation, and automatic control.
Abstract: A stochastic approximation, or stochastic gradient procedure, is a method of estimating the minimum of a criterion function despite the random effects of noise. It was originally conceived by the mathematician as a means of estimating the minimum of a regression function. In recent years, however, stochastic approximation has received considerable attention in the engineering field because of its potential application to learning systems, parameter estimation, and automatic control.
TL;DR: A method of system identification is introduced which involves an operator in an interactive scheme using a graphic display linked to a hybrid computer to obtain a best-fit identification.
Abstract: A method of system identification is introduced which involves an operator in an interactive scheme using a graphic display linked to a hybrid computer. Correlation functions of process input/output data are calculated by the digital computer. A model of the system's dynamics is built on the analogue computer, and the model parameters are adjusted, by visual interaction, to obtain a best-fit identification