• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.8
• /
• pp.39-47
• /
• 1999

This paper describes on-line retrainable RBF network in order to control the coiling temperature for a hot coil strip at Pohang Iron & Steel Company(POSCO). The proposed neural network can be used for improving conventional rule-based lookup table, which generates a heat transmission coefficient. To cope with time-varying characteristics of hot coil process, additional synaptic weights for on-line retraining purposes are introduced to hidden-to-output weights of conventional RBF network. Those weights are locally adjusted to newly incoming test data while preserving old information trained with off-line past data. Hence the effect of catastrophic interference can be greatly alleviated with the proposed network. In addition, rejection scheme is introduced for reliability concerns. From the experimental results applied to the actual process, it is noticed that overall control performance represents about 2.2% increase compared to the conventional one.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.6
• /
• pp.1379-1386
• /
• 2005

For efficient welding it is necessary to maintain stability of the welding process and control the shape of the welding bead. The welding quality can be controlled by monitoring important parameters, such as, the Arc Voltage, Welding Current and Welding Speed during the welding process. Welding systems use either a vision sensor or an Arc sensor, both of which are unable to control these parameters directly. Therefore, it is difficult to obtain necessary bead geometry without automatically controlling the welding parameters through the sensors. In this paper we propose a novel approach using fuzzy logic and neural networks for improving welding qualify and maintaining the desired weld bead shape. Through experiments we demonstrate that the proposed system can be used for real welding processes. The results demonstrate that the system can efficiently estimate the weld bead shape and remove the welding detects.

• Journal of Sensor Science and Technology
• /
• v.10 no.2
• /
• pp.134-141
• /
• 2001

This paper describes a method to steer an autonomous electric vehicle using magnetic fields. Magnets are embeded along the center of the road and a magneto-resistive sensor is mounted beneath the front bumper of the vehicle. As the vehicle moves along the road neural network controller controls the vehicle using measured magnetic field variation. Based on a single magnets modeling equation, we analyzed three dimensional magnetic field distributions of embeded magnets in series on the center of the road and performed a computer simulation using this results. In simulation study, straight and curved road was configured. The steering controller for the vehicle was designed using neural network and experiment was performed on the real embeded magnets using real autonomous electric vehicle. At the experiment we compensated the earth's magnetic fields and showed a good result driving an autonomous vehicle using proposed method.

• Journal of Sensor Science and Technology
• /
• v.12 no.5
• /
• pp.211-217
• /
• 2003

This paper is proposed that roadway recognition performance improvement for autonomous vehicle using magnetic markers that are embedded along the road center and the sensors mounted on a vehicle, and which changing of magnetic field that is measured along with vehicle driving. For Retrenchment of equipment cost, interval of markers is more expensive than existing method. In order to this, This paper is proposed that interval of markers is founded using magnetic field analysis, and which arrangement method of six magnetic sensors and control method of neural network. This paper is carried out magnetic field analysis, the acquiring of the training patterns, the training of the neural network and composition of steering control, and is verified that roadway recognition performance can improve using computer simulation with proposed methods.

• Journal of the Korean Solar Energy Society
• /
• v.25 no.1
• /
• pp.1-10
• /
• 2005

In this paper, 150W photovoltaic system using neural network tracker is proposed, the system designed as the normal line of the solar cell always runs parallel the ray of the sun. This design can minimize the cosine loss of the system output results of solar cell are sensitive to the change of weather and insolation condition don't react rapidly to parameter condition change such as system circumstance and deterioration. To achieve precise operation of photovoltaic tracker system using method of intelligent control, Neural Network is used in the design of the photovoltaic tracker system drive. The control performance of this system drive influenced by the environment parameter such as weather condition and motor parameter variations. we used synchronous motor in this tracker and the experimental results show that the fixing system shows 10,159[Wh] and tracking system shows 12,360[Wh] electricity.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.51 no.4
• /
• pp.188-195
• /
• 2002

We proposed the design of SVC adaptive fuzzy logic controller(AFLC) using Tabu search and neural network. We tuned the gains of input-output variables of fuzzy logic controller(FLC) and weights of neural network using Tabu search. Neural network was used for adaptively tuning the output gain of FLC. The weights of neural network was learned from the back propagation algorithm in real-time. To evaluate the usefulness of AFLC, we applied the proposed method to single-machine infinite system. AFLC showed the better control performance than PD controller and GAFLS[10] for three-phase fault in nominal load which had used when tuning AFLC. To show the robustness of AFLC, we applied the proposed method to disturbances such as three-phase fault in heavy and light load. AFLC showed the better robustness than PD controller and GAFLC[10].

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.751-755
• /
• 1997

This paper presents a new scheme of neural network controller to improve to improve the robustuous of robot manipulator using digital signal processors. Digital processors, DSPs, are micro-processors that are particularly developed for fast numerical computations involving sums and products of variables. Digital version of most advanced control algorithms can be defined as sums and producrs of measured variables, thus it can be programmed and executed through DSPs. In addition, DSPs are as fist in computation as most 32-bit micro-processors and yet at a fraction of their prices. These features make DSPs a viable computational tool in digital implementation of sophisticated controllers. During past decade it was proposed the well-established theorys for the adaptive control of linear systems, but there exits relativly little gensral theoral for the adaptive control of nonlinear systems. Perforating of the proposed controller is illustrated. This paper describes a new approach to the design of adaptive controller and implementation of real-time control for assembling robotic manipulator using digital signal processor. Digital signal processors used in implementing real time adaptive control algorithm are TMS320C50 series made in TI'Co..

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.5
• /
• pp.865-870
• /
• 2006

The cellular neural networks have shown a vast computing power for the image processing in spite of the simplicity of its structure. However, it is impossible to implement the CNN hardware which would require the same enormous amount of cells as that of the pixels involved in the practical large image. In this parer, the $5\times5$ CNN hardware and the pre post processor which can be used for processing the real large image with a time-multiplexing scheme are implemented. The implemented $5\times5$ CNN hardware and pre post processor is applied to the edge detection of $256\times256$ lena image to evaluate the performance. The total number of block. By the time-multiplexing process is about 4,000 blocks and to control pulses are needed to perform the pipelined operation or the each block. By the experimental resorts, the implemented $5\times5$ CNN hardware and pre post processor can be used to the real large image processing.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.9
• /
• pp.723-732
• /
• 2001

In this paper, the neurofuzzy controller of optically guided AGV is proposed to improve the path-tracking performance A differential steered AGV has front-side and rear-side optical sensors, which can identify the guiding path. Due to the discontinuity of measured data in optical sensors, optically guided AGVs break away easily from the guiding path and path-tracking performance is being degraded. Whenever the On/Off signals in the optical sensors are generated discontinuously, the motion errors can be measured and updated. After sensing, the variation of motion errors can be estimated continuously by the dead reckoning method according to left/right wheel angular velocity. We define the estimated contour error as the sum of the measured contour in the sensing error and the estimated variation of contour error after sensing. The neurofuzzy system consists of incorporating fuzzy controller and neural network. The center and width of fuzzy membership functions are adaptively adjusted by back-propagation learning to minimize th estimated contour error. The proposed control system can be compared with the traditional fuzzy control and decision system in their network structure and learning ability. The proposed control strategy is experience through simulated model to check the performance.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.42 no.2 s.302
• /
• pp.39-48
• /
• 2005

This paper proposes a novel approach to recognize nine kinds of motion for a multifunction myoelectric hand, acquiring four channel EMG signals from electrodes placed on the forearm. To analyze EMG with properties of nonstationary signal, time-frequency features are extracted by wavelet packet transform. For dimensionality reduction and nonlinear mapping of the features, we also propose a feature projection composed of PCA and SOFM. The dimensionality reduction by PCA simplifies the structure of the classifier, and reduces processing time for the pattern recognition. The nonlinear mapping by SOFM transforms the PCA-reduced features to a new feature space with high class separability. Finally a multilayer neural network is employed as the pattern classifier. From experimental results, we show that the proposed method enhances the recognition accuracy, and makes it possible to implement a real-time pattern recognition.