• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.26.6-26
• /
• 2001

In this paper we introduce a new algorithm that enables radial basis function network(RBFN) to be structured automatically and guarantees the stability of the RBFN. Because this new algorithm is efficient and also have the advantage of fast computational speed we adopt this algorithm as online learning scheme for uncertain nonlinear dynamical systems. Based on the fact that a 3-layered RBFN can represent a specific nonlinear function reasonably well by linearly combining a set of nonlinear and localized basis functions, we show that this RBFN can identify the nonlinear system very well without knowing the information of the system in advance.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.276-283
• /
• 2007

It has been shown that the structure identification of a neuro-fuzzy model improves their accuracy performances in a various modeling problems. In this paper, we claim that the structure identification of a neuro-fuzzy model can also reduce the degree of inconsistency of its fuzzy rulebase. Thus, the resulting neuro-fuzzy model serves as more like a structured knowledge representation scheme. For this, we briefly review a structure identification method of a neuro-fuzzy model and propose a systematic method to measure inconsistency of a fuzzy rulebase. The proposed method is applied to problems or fuzzy system reproduction and nonlinear system modeling in order to validate our claim.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.10a
• /
• pp.115-120
• /
• 1997

System identification is the task of inferring a mathematical description of a dynamic system from a series of measurements of the system. There are several motives for establishing mathematical descriptions of dynamic systems. Typical applications encompass simulation, prediction, fault diagnostics, and control system design. The paper demonstrates that neural networks can be used effective for the identification of nonlinear dynamical systems. The content of this paper concerns dynamic neural network models, where not all inputs to and outputs from the networks are measurable. Only one model type is treated, the well-known Innovation State Space model(Kalman Predictor). The identification is based only on input/output measurements, so in fact a non-linear Extended Kalman Filter problem is solved. Even for linear models this is a non-linear problem without any assurance of convergence, and in spite of this fact an attempt is made to apply the principles from linear models, an extend them to non-linear models. Computer simulation results reveal that the identification scheme suggested are practically feasible.

• Proceedings of the KIEE Conference
• /
• 1993.11a
• /
• pp.290-292
• /
• 1993

This paper investigates the identification of discrete time nonlinear system using neural networks with two hidden layers. A New learning method of both NNI and NNC is proposed. For control of the dynamical system we use two neural networks, one for identification and the other for control, and proposed NN control system is based on a framework of MRC. We define a closed loop error. In the proposed learning method, the identification error and the closed loop error are utilized to train the NNI, whareas the control error and the closed loop error are used to train the NNC, The simulation results show that the identification and control schemes suggested are practically feasible and effective.

• Journal of Ocean Engineering and Technology
• /
• v.16 no.6
• /
• pp.25-31
• /
• 2002

A system identification method is introduced to increase the prediction accuracy of a ship's maneuverability in PMM test, analysis. To improve the accuracy of linear hydrodynamic coefficients, the analysis techniques of pure sway and yaw tests are developed, and confirmed. In the analysis of sway tests, accuracy to linear hydrodynamic coefficients depends on the frequency of sway motion. To obtain nonlinear hydrodynamic coefficients for large drift angles, a combined yaw test is introduced. Using this system identification method, runs of PMM test can be reduced while retaining sufficient accuracy, compared to the Fourier integration method. Through the comparisons with sea trial results and the Fourier integration method, the accuracy and efficiency of the newly proposed system identification method, based on least square method, has been validated.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.927-930
• /
• 1992

VSS identification approach is based on following concept, i.e. while in sliding motion, the switching of control inputs refects system uncertainites. Therefore, if there exist some operations that make the information form the switiching control inputs be achievable, then the unknown parameters can be actually identification mechanisms which can fully make use of the available information. Two different types of VSS identifiers are taken into consideration. The first type uses adjustable model whose structure is similar to that of identified systems. From the viewpoint of contro, this type of VSS identifiers may be regraded as direct identifier vecause the identified system is handled as an open loop. On the other hand, if the identified system is controlable in the sense of VSS(sliding mode can be generated through chosing control inputs), the second type of VSS identifier, the indirect VSS identifier, can be constructed according to the linerized system strucutre while staying in sliding mode. Therefroe it can be applied to some nonlinear systems which are not linear in parametric space by general identification algorithms, whereas linear in parametric space when sliding mode is existed.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.710-713
• /
• 1995

Hopfield networks have been applied to the problem of linear system identification. In this paper, Hopfield network based parameter identification scheme of non-linear dynamic systems is proposed. Simulation results demonstrate that Hopfield network can be used effectively for the identification of non-linear systems assuming that the system states and their time derivatives are available. Therefore, the proposed scheme can be applied in fault detection and isolation(FDI) and adaptive control of non-linear systems where the Hopfield networks perform on-line identification of system parameters.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.7
• /
• pp.178-187
• /
• 1999

In order to identify a transfer function model with noise, penalty function method has been widely used. In this method, estimation process for possible model parameters from low to higher order proceeds the model identification process. In this study, based on linear estimation method, a new approach unifying the estimation and the identification of ARMAX model is proposed. For the parameter estimation of a transfer function model with noise, linear estimation method by noise separation is suggested instead of nonlinear estimation method. The feasibility of the proposed model identification and estimation method is verified through simulations, namely by applying the method to time series model. In the case of time series model with noise, the proposed method successfully identifies the transfer function model with noise without going through model parameter identification process in advance. A new algorithm effectively achieving model identification and parameter estimation in unified frame has been proposed. This approach is different from the conventional method used for identification of ARMAX model which needs separate parameter estimation and model identification processes. The consistency and the accuracy of the proposed method has been verified through simulations.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.10
• /
• pp.927-932
• /
• 2007

The sophisticated mathematical model is required for the design and the database construction of the advanced flight control system of UAV. In this paper, flight test of KARI's research UAV, often called DURUMI-II, is implemented for the data acquisition from the maneuver flight. The flight path reconstruction is implemented to ensure that the measured data is consistent and error free. The nonlinear system identification for the refined mathematical modeling is implemented with the verified measurements from the flight path reconstruction. The simulation with the identified results have a good validation when the simulated responses were compared to the flight tested data.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.141-144
• /
• 2003

A composite adaptive dual fuzzy controller combining the approximate mathematical model, linguistic model description, linguistic control rules and identification modeling error into a single adaptive fuzzy controller is developed for a nonlinear system. It ensures the system output tracks the desired reference value and excites the plant sufficiently for accelerating the parameter estimation process so that the control performances are greatly improved. Using the Lyapunov synthesis approach, proposed controller is analyzed and simulation results verify the effectiveness of the proposed control algorithm.