Smart System Research Group

Tenan, South Korea

Smart System Research Group

Tenan, South Korea
SEARCH FILTERS
Time filter
Source Type

Kim H.-T.,Smart System Research Group | Lee K.-W.,Smart System Research Group | Yang H.-J.,Korea Polytechnic University | Kim S.-C.,AmTechnology
International Journal of Precision Engineering and Manufacturing | Year: 2013

We propose a self-learning method for automatic wafer alignment in the semiconductor manufacturing process. A feed forward neural network is trained by and used for wafer alignment. The network determines the movement of kinematic parts from the misalignment inspected by machine vision. However, it is time-consuming and inconvenient to obtain training data in this way. So, we built an automatic learning rule to gather the data and train the network. The network may determine wrong outputs and cause other misalignments at first, but the error can decrease as the training proceeds. The training sets consisted of a variation of misalignment data and the movement of an alignment stage. Five recent sets are used for training and others are dismissed or forgotten. This retrained network tried aligning, measured misalignment, and made new training sets. This sequence makes it possible to acquire alignment skill and automate the process. After learning, automatic alignment accomplished sub-pixel accuracy for several cases of misalignment. The result showed that the proposed method could be applied to the semiconductor manufacturing process. Its performance improved about 6% compared with conventional algorithms. © 2013 Korean Society for Precision Engineering and Springer-Verlag Berlin Heidelberg.


Kim H.-T.,Smart System Research Group | Kim C.-H.,Smart System Research Group | Choi S.-B.,Inha University | Moon S.-J.,Shock and Vibration Team | Song W.-G.,Industrial Technology Group
Smart Materials and Structures | Year: 2014

With the goal of vibration control and isolation in a clean room, we propose a new type of air mount which consists of pneumatic, electromagnetic (EM), and magnetorheological (MR) actuators. The air mount is installed below a semiconductor manufacturing machine to reduce the adverse effects caused by unwanted vibration. The proposed mechanism integrates the forces in a parallel connection of the three actuators. The MR part is designed to operate in an air spring in which the EM part is installed. The control logic is developed with a classical method and a switching mode to avoid operational mismatch among the forces developed. Based on extended microprocessors, a portable, embedded controller is installed to execute both nonlinear logic and digital communication with the peripherals. The pneumatic forces constantly support the heavy weight of an upper structure and maintain the level of the air mount. The MR damper handles the transient response, while the EM controller reduces the resonance response, which is switched mutually with a threshold. Vibration is detected by laser displacement sensors which have submicron resolution. The impact test results of three tons load weight demonstrate practical feasibility by showing that the proposed triple-actuating mechanism can reduce the transient response as well as the resonance in the air mount, resulting in accurate motion of the semiconductor manufacturing machine. © 2014 IOP Publishing Ltd.


Kim H.,Smart System Research Group | Kim C.,Smart System Research Group | Choi S.,Inha University | Moon S.,Shock and Vibration Team | Song W.,Structure Dynamic Design Group
19th International Congress on Sound and Vibration 2012, ICSV 2012 | Year: 2012

For the purpose of vibration control and isolation in a clean room, we propose an active device combining pneumatic, electro-magnetic (EM) and magneto-rheological (MR) forces. The mechanism was the integration of the forces in a parallel connection. The MR parts are designed to operate in the air spring which the EM parts border out side. The control logic was organized with classical methods and switching mode for avoiding operational mismatch among the forces. PC-based controller was constructed to execute the non-linear logic and communication with the peripherals. Pneumatics was constantly operating for maintaining a level of the air mount. MR handles transient response and EM control undertook reduction of resonance response, which were switched mutually with a threshold. The vibration was detected by a laser displacement sensor which had sub-micron resolution. Four active devices were constructed and installed under 3 tons of air mount. The impact test results show the possibility that the proposed device can reduce the transient response as well as the resonance in the air mount.


Kim H.,Smart System Research Group | Kim C.,Smart System Research Group | Kang S.,Smart System Research Group | Moon S.,Shock and Vibration Team | Lee G.,Structure Dynamic Design Group
19th International Congress on Sound and Vibration 2012, ICSV 2012 | Year: 2012

An active device using electro-magnetic forces was constructed and experimented for the purpose of minimizing resonance in air mounts of clean rooms. The air mounts are vulnerable to low frequency resonance due to heavy weight and low stiffness. We developed and tested a hybrid structure of the active device which is composed of pneumatic and electro-magnetic parts. The pneumatic parts in the device support heavy weights under air mounts and the electro-magnetic parts reduce the resonance. The electro-magnetic parts are composed of dual stators and an armature which surround the pneumatic parts. The resonance can decrease when the electro-magnetic forces are generated in gaps between the stators and the armature. Four active devices were installed under a 3-ton surface plate for vibration test apparatus. The vibration was detected by eddy-current sensors. Discrete P Control logic was based on displacement and embedded in a C6713 DSP. The results from impact tests show that the peak magnitude in the resonance frequency can be reduced to-10dB.


Kim H.T.,Smart System Research Group | Kim S.T.,Smart System Research Group | Cho Y.J.,Smart System Research Group
International Journal of Optomechatronics | Year: 2012

We considered the issue of color mixing from the viewpoint of gray scale images and investigated the effect of the mixed light. The goal of this study was to improve the quality of the grey scale images by using mixed RGB illumination. The mixing mechanism and image quality were formulized and applied to a machine vision system with an RGB mixer. The RGB mixer was made of RGB LEDs, a power AMP, a mixing chamber, and a bundle of optical fibers. The intensity of the RGB light was adjusted by varying the 12 bit DAC voltage. The optical characteristics of the RGB LEDs were measured by a spectrometer. The color variation of the mixed light was monitored by a chromatic meter. Grey scale images were captured and analyzed under various lighting conditions provided by the mixer. The optimal illumination was determined by the distribution of the grey scale in the captured image. The chromatic diagram showed that various colors can be generated, and the image histogram showed that the mixed light can increase the quality of the grey scale image. © 2012 Copyright Taylor and Francis Group, LLC.


Kim H.,Smart System Research Group | Kim S.,Smart System Research Group | Cho Y.,Smart System Research Group
2012 International Symposium on Optomechatronic Technologies, ISOT 2012 | Year: 2012

Optimum searching methods were applied to find optimal illumination of a single color to enhance the quality of a monochrome image and reduce searching time. The problem was defined as 1D searching optimization between light intensity and image sharpness. The image sharpness is the degree of distinctness and is a non-linear function of the input voltage, which can be used to adjust the light intensity. We considered conventional optimum search methods, such as steepest descent, conjugate gradient, Newton's method, bisection, and golden section. These derivative methods and direct searches were tested for a sample pattern using single color lights under coaxial illumination. The iteration for the optimal condition was 6.7% of full scanning in average. The test results show that the steepest descent and golden search are recommended. We checked the possibility of applying these search methods to automatic lighting in machine vision. © 2012 IEEE.


Kim H.,Smart System Research Group | Kim S.,Smart System Research Group | Kim J.,Smart System Research Group
2012 International Symposium on Optomechatronic Technologies, ISOT 2012 | Year: 2012

Quick search for optimal illumination using mixed color is proposed, and the possibility for use in industrial vision systems is considered. Multidimensional formulation of the optimum search methods is reviewed to use multiple inputs for various color sources and one output of image sharpness. The voltage inputs are used to adjust the intensity of each color source and the optimal illumination is found by the maximum of the sharpness. The optimum search methods reveal that the number of iterations for the optimal condition in mixed color illumination was under 0.1% the number required by equal section search. The searching paths and convergence were considered in both of the input coordinate and the color coordinate. The results show that the optimum search methods make it possible to find the optimal lighting conditions quickly and accurately. © 2012 IEEE.


Kim H.,Smart System Research Group | Kim S.,Smart System Research Group | Kim J.,Smart System Research Group
International Journal of Optomechatronics | Year: 2013

Quick search for optimal illumination using mixed color is proposed, and the possibility for use in industrial vision systems is considered. Multidimensional formulation of the optimum search methods is reviewed to use multiple inputs for various color sources and one output of image sharpness. The voltage inputs are used to adjust the intensity of each color source and the optimal illumination is found by the maximum of the sharpness. The optimum search methods reveal that the number of iterations for the optimal condition in mixed color illumination was under 0.1% the number required by equal section search. The searching paths and convergence were considered in both of the input coordinate and the color coordinate. The results show that the optimum search methods make it possible to find the optimal lighting conditions quickly and accurately. © 2013 Copyright Taylor and Francis Group, LLC.


Kim H.T.,Smart System Research Group | Jin K.C.,Smart System Research Group
Applied Mechanics and Materials | Year: 2013

Recent mechatronic systems, such as inspection machines or 3D imaging apparatuses, acquire and compute massive data for final results. A host in the mechatronic system is commonly composed of multiple hardware devices which interface with high-speed external signals. The host and the devices usually have large memory, so efficient data management is important due to data storage and transfer. In our software structure, each device is managed by respective application and large shared memory (LSM) is allocated in the host for the massive data. The shared memory is accessible from the device applications. Actions of the mechatronic system are driven by combining and broadcasting events through and inter-process communication (IPC). The model with LSM and IPC was applied to a 3D RF imaging system. We expect the proposed model can also be applied to machine vision with big image and engineering simulation with hardware accelerators. © (2013) Trans Tech Publications, Switzerland.


Jin K.C.,Smart System Research Group | Kim H.T.,Smart System Research Group
Applied Mechanics and Materials | Year: 2013

Mojette Transform (MOT) is used mainly in imaging implementation of mechatronicbased imaging system to reconstruct a discrete signal from a finite set of projection planes. The MOT uses a specific algorithm, called Corner Based Inversion (CBI), to reconstruct an image from its projections offering high-speed computing properties. Moreover, the MOT ensures a very low complexity in comparison to the reconstruction based on Fast Fourier Transform (FFT). In this paper, Graphic Processing Unit (GPU) based MOT is presented and also CPU and GPU processing are issued from 1283 image pixels. In the result, performance differences between the CPU and GPU architectures are discussed, and an approach of fast improvement in architectural efficiency is recommend. © (2013) Trans Tech Publications, Switzerland.

Loading Smart System Research Group collaborators
Loading Smart System Research Group collaborators