• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.09a
• /
• pp.325-333
• /
• 1995

Back-propagation is the most widely used algorithm for supervised learning in multi-layer feed-forward networks. However, back-propagation is very slow in convergence. In this paper, a new weight initialization method, called rough map initialization, in multi-layer perceptrons is proposed. To overcome the long convergence time, possibly due to the random initialization of the weights of the existing multi-layer perceptrons, the rough map initialization method initialize weights by utilizing relationship of input-output features with singular value decomposition technique. The results of this initialization procedure are compared to random initialization procedure in encoder problems and xor problems.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.1
• /
• pp.1-9
• /
• 2006

This paper presents the modeling and multivariable feedback control of a novel high-precision multi-dimensional positioning stage. This integrated 6-degree-of-freedom. (DOF) motion stage is levitated by three aerostatic bearings and actuated by 3 three-phase synchronous permanent-magnet planar motors (SPMPMs). It can generate all 6-DOF motions with only a single moving part. With the DQ decomposition theory, this positioning stage is modeled as a multi-input multi-output (MIMO) electromechanical system with six inputs (currents) and six outputs (displacements). To achieve high-precision positioning capability, discrete-time integrator-augmented linear-quadratic-regulator (LQR) and reduced-order linearquadratic-Gaussian (LQG) control methodologies are applied. Digital multivariable controllers are designed and implemented on the positioning system, and experimental results are also presented in this paper to demonstrate the stage's dynamic performance.

• International Journal of Internet, Broadcasting and Communication
• /
• v.4 no.2
• /
• pp.1-6
• /
• 2012

Control Theory for continuous-time system has been well developed. Due to the development of computer technology, digital control scheme are employed in many areas. When delays are in control systems, it is hard to control the system efficiently. Delays by controller-to-actuator and sensor-to-controller deteriorate control performance and could possibly destabilize the overall system. In this paper, a new approximated discretization method and digital design for control systems with multiple state, input and output delays and a generalized bilinear transformation method with a tunable parameter are also provided, which can re-transform the integer time-delayed discrete-time model to its continuous-time model. Illustrative examples are given to demonstrate the effectiveness of the developed method.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.381-385
• /
• 1992

In this paper, dynamic modeling and adaptive control problems for a space robot system are discussed. The space robot consist of a robot manipulator mounted on a free-floating base where no attitude control is applied. Using an extended robot model, the entire space robot can be viewed as an under-actuated robot system. Based on nonlinear control theory, the extended space robot model can then be decomposed into two subsystems: one is input-output exactly linearizable, and the other is unlinearizable and represents an internal dynamics. With this decomposition, a normal form-augmentation approach and an augmented state-feedback control are proposed to facilitate the design of adaptive control for the space robot system against parameter uncertainty, unknown dynamics and unmodeled payload in space applications. We demonstrate that under certain conditions, the entire space robot can be represented as a full-actuated robot system to avoid the inclusion of internal dynamics. Based on the dynamic model, we propose an adaptive control scheme using Cartesian space representation and demonstrate its validity and design procedure by a simulation study.

• ETRI Journal
• /
• v.42 no.3
• /
• pp.333-340
• /
• 2020

We consider a hybrid combiner design for downlink massive multiple-input multiple-output systems when there is residual inter-user interference and each user is equipped with a limited number of radio frequency (RF) chains (less than the number of receive antennas). We propose a hybrid combiner that minimizes the mean-squared error (MSE) between the information symbols and the ones estimated with a constant amplitude constraint on the RF combiner. In the proposed scheme, an iterative alternating optimization method is utilized. At each iteration, one of the analog RF and digital baseband combining matrices is updated to minimize the MSE by fixing the other matrix without considering the constant amplitude constraint. Then, the other matrix is updated by changing the roles of the two matrices. Each element in the RF combining matrix is obtained from the phase component of the solution matrix of the optimization problem for the RF combining matrix. Simulation results show that the proposed scheme performs better than conventional matrix-decomposition schemes.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.348-351
• /
• 1995

This note proposes a robust LQR method for systems with structured real parameter uncertainty based on Riccati equation approach. Emphasis is on the reduction of design conservatism in the sense of quadratic performance by utilizing the uncertainty structure. The class of uncertainty treated includes all the form of additive real parameter uncertainty, which has the multiple rank structure. To handle the structure of uncertainty, the scaling matrix with block diagonal structure is introduced. By changing the scaling matrix, all the possible set of uncertainty structures can be represented. Modified algebraic Riccati equation (MARE) is newly proposed to obtain a robust feedback control law, which makes the quadratic cost finite for an arbitrary scaling matrix. The remaining design freedom, that is, the scaling matrix is used for minimizing the upper bound of the quadratic cost for all possible set of uncertainties within the given bounds. A design example is shown to demonstrate the simplicity and the effectiveness of proposed method.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1775-1776
• /
• 2008

This paper presents a adaptive fuzzy control suitable for motion control of multi-link robot manipulators with uncertainties. When joint velocities are available, full state adaptive fuzzy feedback control is designed to ensure the stability of the closed loop dynamic. If the joint velocities are not measurable, an observer is introduced and an adaptive output feedback control is designed based on the estimated velocities. To reduce the number of fuzzy rules of the fuzzy controller, we consider the properties of robot dynamics and the decomposition of the unknown input gain matrix. The proposed controller is robust against uncertainties and external disturbances. The validity of the control scheme is demonstrated by computer simulations on a two-link robot manipulator.

• Proceedings of the KSME Conference
• /
• 2000.04a
• /
• pp.566-572
• /
• 2000

The System identification is the process of developing or improving a mathematical model of a physical system using experimental data of the input, output and noise relationship. The field of system identification has been an important discipline within the automatic control area. The reason is the requirement that mathematical models having a specified accuracy must be used to apply modem control methods. In this paper, it is confirmed that we can obtain transfer function of flexible beam that is expressed in the forms of identified state-space system matrix A, B, C, D and identified observer gain G using Eigensystem Realization Algorithm including singular value decomposition. And these matrices can be applied to the automatic control. In addition to, it is also confirmed that transfer function can express a system using identified observer gain G, in spite of a noisy data or a periodic disturbance.

• Journal of information and communication convergence engineering
• /
• v.4 no.2
• /
• pp.75-78
• /
• 2006

This work presents a low complexity maximum-likelihood decoder for signal detection in VBLAST-STBC system, which employs non-square O-STBC code rate 3/4. Stacking received symbols from different symbol duration and applying QR decomposition result in the special format of upper triangular matrix R so that the proposed decoder is able to provide not only ML-like BER performance but also very low computational load. The low computational load and ML-like BER performance properties of the proposed decoder are verified by computer simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.7
• /
• pp.2879-2890
• /
• 2020

In this paper, we study the estimation of angle, range and polarization parameters of a bistatic polarization sensitive frequency diverse array multiple-input multiple-output (PSFDA-MIMO) radar system. The application of polarization sensitive array in receiver is explored. A signal model of bistatic PSFDA-MIMO radar system is established. In order to utilize the multi-dimensional structure of array signals, the matched filtering radar data can be represented by a third-order tensor model. A joint estimation of the direction-of-departure (DOD), direction-of-arrival (DOA), range and polarization parameters based on parallel factor (PARAFAC) algorithm is proposed. The proposed algorithm does not need to search spectral peaks and singular value decomposition, and can obtain automatic pairing estimation. The method was compared with the existing methods, and the results show that the performance of the method is better. Therefore, the accuracy of the parameter estimation is further improved.