• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.39 no.4
• /
• pp.414-422
• /
• 2002

In this paper, we propose a new method of adaptive neuro-fuzzy control using CFCM(Conditional Fuzzy c-means) clustering and fuzzy equalization method to deal with adaptive control problem. First, in the off-line design, CFCM clustering performs structure identification of adaptive neuro-fuzzy control with the homogeneous properties of the given input and output data. The parameter identification are established by hybrid learning using back-propagation algorithm and RLSE(Recursive Least Square Estimate). In the on-line design, the premise and consequent parameters are tuned to RLSE with forgetting factor due to a characteristic of time variant. Finally, we applied the proposed method to the water temperature control system and obtained better results than previous works such as fuzzy control.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.5
• /
• pp.858-865
• /
• 2008

In this paper, Capacitive Telemetry RF Sensor System using Recursive Least Square (RLS) algorithm was proposed. General Telemetry RF Sensor System means that it should be "wireless", "implantable" and "batterless". Conventional Telemetry RF Sensor System adopts Integrated Circuit type, but there are many defects like complexity of structure and the limitation of large power consumption in some cases. In order to overcome these disadvantages, Telemetry RF Sensor System based on inductive coupling principle was proposed in this paper. Proposed Telemetry RF Sensor System is very simple because it consists of R, L and C and measures the changes of environment like pressure and humidity in the type of capacitive value. This system adopted RLS algorithm for estimation of this capacitive parameter. For the purpose of applying RLS algorithm, proposed system was mathematically modelled with phasor method and was quasi-linearized. As two parameters such as phase and amplitude of output voltage for estimation were needed, Phase Difference Detector and Amplitude Detector were proposed respectively which were implemented using TMS320C2812 made by Texas Instrument. Finally, It is verified that the capacitance of proposed telemetry RF Sensor System using RLS algorithm can be estimated efficiently under noisy environment.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2002.11a
• /
• pp.573-577
• /
• 2002

A new estimated neural networks are proposed in order to measure nonlinear road environments in realtime. This new neural networks is Error Estimated Neural Networks. The structure of it is similar to recurrent neural networks; a delayed output as the input and a delayed error between the output of plant and neural networks as a bias input. In addition, we compute the desired value of hidden layer by an optimal method instead of transfering desired values by backpropagation and each weights are updated by RLS(Recursive Least Square). Consequently, this neural networks are not sensitive to initial weights and a learning rate, and have a faster convergence rate than conventional neural networks. We can estimate nonlinear models in realtime by the proposed networks and control nonlinear models. To show the performance of this one, we control 7 degree simulation, this controller and driver were proved to be effective to drive a car in the environments of nonlinear road systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.505-510
• /
• 2005

The RPO(randomized policy optimizer) algorithm, which utilizes probabilistic policy for the action selection, is a recently developed tool in the area of reinforcement learning, and has been shown to be very successful in several application problems. In this paper, we propose a modified RPO algorithm, whose critic network is adapted via RLS(Recursive Least Square) algorithm. In order to illustrate the applicability of the modified RPO method, we applied the modified algorithm to Kimura's robot and observed very good performance. We also developed a MATLAB-based animation program, by which the effectiveness of the training algorithms on the acceleration or the robot movement were observed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.8
• /
• pp.754-759
• /
• 2001

In this paper, we propose a TSK(Takagi-Sugeno-Kang)-type fuzzy classifier using PCA(Principal Component Analysis), FCM(Fuzzy c-Means) clustering, ANFIS(Adaptive Neuro-Fuzzy Inference System) and hybrid GA(Genetic Algorithm). First, input data is transformed to reduce correlation among the data components by PCA. FCM clustering is applied to obtain a initial TSK-type fuzzy classifier. Parameter identification is performed by AGA(Adaptive GA) and RLSE(Recursive Least Square Estimate). Finally, we applied the proposed method to Iris data classificationl problems and obtained a better performance than previous works.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.22 no.1
• /
• pp.118-123
• /
• 2008

This paper proposes a frequency transfer function synthesis for approximating a high-order model with resonance to a low-order model in the frequency domain. The presented model approximation method is based on minimizing the error function weighted by the numerator polynomial of approximated models, which is used of the RLS(Recursive Least Square) technique to estimate the coefficient vector of approximated models. The proposed method provides better fitting in a low frequency and peak resonance. And an example is given to illustrate feasibilities of the suggested schemes.

• Journal of Auto-vehicle Safety Association
• /
• v.5 no.1
• /
• pp.62-68
• /
• 2013

This paper describes development of 4 Wheel Drive (4WD) Electric Vehicle (EV) based driving control algorithm for severe driving situation such as icy road or disturbance. The proposed control algorithm consists three parts : a supervisory controller, an upper-level controller and optimal torque vectoring controller. The supervisory controller determines desired dynamics with cornering stiffness estimator using recursive least square. The upper-level controller determines longitudinal force and yaw moment using sliding mode control. The yaw moment, particularly, is calculated by integration of a side-slip angle and yaw rate for the performance and robustness benefits. The optimal torque vectoring controller determines the optimal torques each wheel using control allocation method. The numerical simulation studies have been conducted to evaluated the proposed driving control algorithm. It has been shown from simulation studies that vehicle maneuverability and lateral stability performance can be significantly improved by the proposed driving controller in severe driving situations.

• Wind and Structures
• /
• v.11 no.1
• /
• pp.19-33
• /
• 2008

This research investigates the adaptive input estimation method applied to the multilayer shearing stress structure. This method is to estimate the values of wind load inputs by analyzing the active reaction of the system. The Kalman filter without the input term and the adaptive weighted recursive least square estimator are two main portions of this method. The innovation vector can be produced by the Kalman filter, and be applied to the adaptive weighted recursive least square estimator to estimate the wind load input over time. This combined method can effectively estimate the wind loads to the structure system to enhance the reliability of the system active performance analysis. The forms of the simulated inputs (loads) in this paper include the periodic sinusoidal wave, the decaying exponent, the random combination of the sinusoidal wave and the decaying exponent, etc. The active reaction computed plus the simulation error is regard as the simulated measurement and is applied to the input estimation algorithm to implement the numerical simulation of the inverse input estimation process. The availability and the precision of the input estimation method proposed in this research can be verified by comparing the actual value and the one obtained by numerical simulation.

• Journal of the Korean Society for Precision Engineering
• /
• v.11 no.1
• /
• pp.138-149
• /
• 1994

This paper introduces a new monitoring technique which utilizes an adaptive signal processing for feature generation, coupled with a multilayered merual network for pattern recognition. The cutting force signal in face milling operation was modeled by a low order discrete autoregressive model, shere parameters were estimated recursively at each sampling instant using a parameter adaptation algorithm based on an RLS(recursive least square) method with discounted measurements. The influences of the adaptation algorithm parameters as well as some considerations for modeling on the estimation results are discussed. The sensitivity of the extimated model parameters to the tool state(new and worn tool)is presented, and the application of a multilayered neural network to tool state monitoring using the previously generated features is also demonstrated with a high success rate. The methodology turned out to be quite suitable for in-process tool wear monitoring in the sense that the model parameters are effective as tool state features in milling operation and that the classifier successfully maps the sensors data to correct output decision.

• Journal of the Korea Computer Industry Society
• /
• v.8 no.4
• /
• pp.263-268
• /
• 2007

It applied the linear tapped delay line structure to the least mean square algorithm and the recursive least square algorithm it investigated the mean square error characteristics of compensation algorithm. The purpose of this paper is to propose multi-tap update algorithm, which is superior to compensation capacity of data, and then compare and analyze it from the perspective of convergence characteristics at time invariant transmission channel and frequency selective fading channel.