Earticle

음성 인식 시스템은 부정확한 음성 신호의 입력으로 특징을 추출하여 인식할 경우 오인식의 결과가 나타나 거나 유사한 음소로 인식된다. 따라서 본 논문에서는 음소가 갖는 특징을 기반으로 음소 유사율과 신뢰도 측정을 이 용한 음성 인식 오류 보정 방법을 제안하였다. 음소 유사율은 학습 모델의 음소에 MFCC와 LPC 특징 추출 방법을 이용하여 구하였으며 신뢰도로 측정하였다. 음소 유사율과 신뢰도를 측정하여 오인식되는 오류를 최소화하였으며 음 성 인식 과정에서 오류로 판명된 음성에 대하여 오류 보정을 수행하였다. 본 논문에서 제안한 시스템을 적용한 결과 98.3%의 인식률과 95.5%의 오류 보정율을 나타내었다.

Speech recognition system is input of inaccurate vocabulary by feature extraction case of recognition by appear result of unrecognized or similar phoneme recognized. Therefore, in this paper, we propose a speech recognition error correction method using phoneme similarity rate and reliability measures based on the characteristics of the phonemes. Phonemes similarity rate was phoneme of learning model obtained used MFCC and LPC feature extraction method, measured with reliability rate. Minimize the error to be unrecognized by measuring the rate of similar phonemes and reliability. Turned out to error speech in the process of speech recognition was error compensation performed. In this paper, the result of applying the proposed system showed a recognition rate of 98.3%, error compensation rate 95.5% in the speech recognition.

To compensate for the motion errors in hydrostatic tables, a method to actively control the clearance of a bearing corresponding to the amount of error using actively controlled capillaries is introduced in this paper. The design method for an actively controlled capillary that considers the output rate of a piezo actuator and the amount of error that must be corrected is described. The basic characteristics of such a system were tested, such as the maximum controllable range of the error, micro-step response, and available dynamic bandwidth when the capillary was installed in a hydrostatic table. The tests demonstrated that the maximum controllable range was $2.4\;{\mu}m$, the resolution was 27 nm, and the frequency bandwidth was 5.5 Hz. Simultaneous compensation of the linear and angular motion errors using two actively controlled capillaries was also performed for a hydrostatic table driven by a ballscrew and a DC servomotor. An iterative compensation method was applied to improve the compensation characteristics. Experimental results showed that the linear and angular motion errors were improved to $0.12{\mu}m$ and 0.20 arcsec, which were about $1/15^{th}$ and $1/6^{th}$ of the initial motion errors, respectively. These results confirmed that the proposed compensation method improves the motion accuracy of hydrostatic tables very effectively.

A precision four-degree-of-freedom measurement system has been developed for simultaneous measurement of four motion errors of a linear stage, which include straightness and angular errors, The system employs a retro-reflector to detect the straightness errors and a plane mirror to detect the angular errors. A common-path compensation method for laser beam drift is put forward, and the experimental results show that the influences of beam drift on four motion errors can be reduced simultaneously. In comparison with the API 5D laser measuring system, the accuracy for straightness measurement is about ${\pm}1.5{\mu}m$ within the measuring range of ${\pm}650{\mu}m$, and the accuracy for pitch and yaw measurements is about ${\pm}1.5$ arc-seconds within the range of ${\pm}600$ arc-seconds.

This study proposes the compensation method for the mechanical deflection error of a SCARA robot. While most studies on the related subject have dealt with the development of a control algorithm for improvement of robot accuracy, this study presents the control method reflecting the mechanical deflection error which is predicted in advance. The deflection at the end of the gripper of SCARA robot is caused by the self-weights and payloads of Arm 1, Arm 2 and quill. If the deflection is constant even though robot’ posture and payload vary, there may not be a big problem on robot accuracy because repetitive accuracy, that is relative accuracy, is more important than absolute accuracy in robot. The deflection in the end of the gripper varies as robot’ posture and payload change. That’ why the moments ,  and  working on every joint of a robot vary with robot’ posture and payload size. This study suggests the compensation method which predicts the deflection in advance with the variations in robot’ posture and payload using neural network. To do this, I chose the posture of robot and the payloads at random, found the deflections by the FEM analysis, and then on the basis of this data, made compensation possible by predicting deflections in advance successively with the variations in robot’ posture and payload through neural network learning.

This paper proposes a new model to compensate for errors of a five-axis machine tool. A matrix with error components, that is, an error matrix, is separated from the error synthesis model of a five-axis machine tool. Based on the kinematics and inversion of the error matrix which can be obtained not by using a numerical method, an error compensation model is established and used to calculate compensation values of joint variables. The proposed compensation model does not need numerical methods to find the compensation values from the error compensation model, which includes nonlinear equations. An experiment using a double ball-bar is implemented to verify the proposed model.

The object of the thermal error compensation system in machine tools is improving the accuracy of a machine tool through real time error compensation. The accuracy of the machine tool totally depends on the accuracy of thermal error model. A thermal error model can be obtained by appropriate combination of temperature variables. The proposed method for optimal variable selection in the thermal error model is based on correlation grouping and successive regression analysis. Collinearity matter is improved with the correlation grouping and the judgment function which minimizes residual mean square is used. The linear model is more robust against measurement noises than an engineering judgement model that includes the higher order terms of variables. The proposed method is more effective for the applications in real time error compensation because of the reduction in computational time, sufficient model accuracy, and the robustness.

This paper is concerned with the development of Internet-based robot system controlled on the remote site via the Internet. In order to draw the public attention into this exciting system, we built the simple system by which a robot is moved in response to answer for the given OX quizzes. As the primary research fer Internet-based robot control, this study focuses on the development of user-friendly interface by which a beginner achieves information for a robot on the remote site from the 3D virtual simulator and the real camera image. for the compensation of Internet time delay, position prediction simulator is implemented in the user interface.

Robot calibration is very important to improve the accuracy of robot manipulators. However, the calibration procedure is very time consuming and laborious work for users. In this paper, we propose a method of relative error compensation to make the calibration procedure easier. The method is completed by a Pi-Sigma network architecture which has sufficient capability to approximate the relative relationship between the accuracy compensations and robot configurations while maintaining an efficient network learning ability. By experiment of 4-DOF SCARA robot, KIRO-3, it is shown that both the error of joint angles and the positioning error of end effector are drop to 15$\%$. These results are similar to those of other calibration methods, but the number of measurement is remarkably decreased by the suggested compensation method.

A volumetric error compensation method was stueide with measuring systematic error of a Coordinate Measuring Machine(CMM). The volumetric error equations were proposed for a Moving Bridge type CMM. Using the error equations, error vectors in the measuring volume were corrected by a software method. The CMM was controlled by the compensation program separately in the measuring and moving function of the CMM proving. The linear accuracy of the CMM was measured by the Laser Interferometer and compared with the data before the volumetric error compensation. This method was proved as low cost and effective to reduce the systematic error of the CMM, as no hardware modification is required.

In this paper, we studied a vibration problem that is critical and common to most precision measurement systems. For micro mechanical part measurements, results obtained from the vision-based precision measurement system may contain errors due to the vibration. In order to defeat this generic problem, for the current study, a PC based image processing technique was used first, to assess the effect of the vibration to the precision measurement and second, to develop an in-situ calibration algorithm that automatically compensate the measurement results in real time. We used a set of stereoscopic CCD cameras to acquire the images for the dimensional measurement and the reference measurement. The mapping function was obtained through the in-situ calibration to compensate the measurement results and the statistical analysis for the actual results is provided in the paper. Based on the current statistical study, it is expected to obtain high precision results for the micro measurement systems.

A new hardware compensation method reducing displacement measurement errors, caused by tilt of index scale in moire linear encoders, has been developed. In conventional moire linear encoders, the detectors are aligned perpendicular to the line of moire fringes this structure is very sensitive to an unwanted tilt of the gratings. In this paper, a newly designed grating, called a phase-modulated grating, is developed to compensate for non-orthogonal errors. By using the phase-modulated grating instead of a conventional index, it is possible to reduce non-orthogonal errors of moire linear encoders.

The online realtime positioning error compensation system 'SKY-PACS' is developed to correct geometric errors, thermal errors and tool deflection errors induced by cutting forces on the vertical machining center. 'SKY-PACS' communicates position commands and position compensation signals with the CNC controller at 100Hz, which is CNC control frequency. So the compensation procedure can be applied during axis movement. Using 'SKY-PACS', Maximum 1 axis positioning accuracy was corrected from 5{\mu}m$ to 2{\mu}m$and the squareness error of X-Y table was corrected from 51{\mu}m$/m to below 4{\mu}m$/m. The error compensation under the cutting condition is carried out by ISO10791-7. And the measurement of test-pieces shows that the roundness is corrected rom 8{\mu}m$ to below 5{\mu}m$.

Micro laser fabrication techniques can potentially be used for the manufacture of microstructures on the thin flat surfaces with large diameter that are frequently used in semiconductor industries. However, the large size of wafers can cause the degraded machining accuracy of the surface because it can be tilted or distorted by geometric errors of machines or the holding fixtures, etc. To overcome these errors the off-line focusing error compensation method is proposed. By using confocal autofocus system, the focusing error profile of machined surface is measured along the pre-determined path and can be compensated at the next machining process by making the corrected motion trajectories. The experimental results for silicon wafers and invar flat surfaces show that the proposed method can compensate the focusing error within the level of below $6.9{\mu}m$ that is the depth of focus required for the laser micromachining process.

The major role of temperature sensors in thermal error compensation system of machine tools is improving machining accuracy by supplying reliable temperature data on the machine structure. This paper presents a new method for fault diagnosis of temperature sensors and recovery of faulted data to establish the reliability of thermal error compensation system. The detection of fault and its location is based on the correlation coefficients among temperature data from the sensors. The multiple linear regression model which is prepared using complete normal data is also used fur the recovery of faulted data. The effectiveness of this method was tested by comparing the computer simulation results and measured data in a CNC machining center.

This paper suggests a methodology fur improving the machining accuracy by compensating for the machining errors based on on-machine measurement process. Probing errors and machine tool errors included in the measurement data were calibrated or compensated to obtain the actual machining errors. Machine tool errors were modeled in forward and backward directions according to the axis movement direction to consider the effects of backlash errors on the measurement data, and model parameters were determined by measuring a cube array artifact. A rectangular workpiece was machined and then measured with a touch probe as a verification experiment. Machining experiments showed that the machining errors were reduced to within the designated tolerance after compensating for the actual machining errors by modifying the original footpath for the next-step machining.

A fine stage is developed for the 3-DOF error compensation of a linear axis in order to improve the positioning accuracy. This stage is designed as a planar parallel mechanism, and the joints are based on a flexure hinge to achieve ultra-precise positioning. Also, the effect of Abbe's offsets between the measuring and driving coordinate systems is minimized to ensure an exact error compensation. The mode shapes of the designed stage are analyzed to verify the desired 3-DOF motions, and the workspace and displacement of a piezoelectric actuator (PZT) for compensation are analyzed using forward and inverse kinematics. The 3-DOF error of a linear axis is measured and compensated by using the developed fine stage. A marked improvement is observed compared to the results obtained without error compensation. The peak-to-valley (PV) values of the positional and rotational errors are reduced by 92.6% and 91.3%, respectively.

Machine tools are engineered to give high dimensional accuracy in machining operation. However, errors due to thermal effects degrade dimensional accuracy of machine tools considerably, and many machine tools are equipped with thermal error compensation function. In general, thermal errors can be generated in the angular directions as well as linear directions. Among them, thermal errors in the angular directions contribute a large amount of error components in the presence of offset distance as in the case of Abbe error. Because most of thermal error compensation function is based on a good correlation between temperature change and thermal deformation, angular thermal deformation is often to be the most difficult hurdle for enhancing compensation accuracy. In this regard, this paper investigates the effect of thermal bending to total thermal error and gives how to deal with thermally induced bending effects in thermal error compensation.

SLA (Stereolithography Apparatus) it a process used to rapidly produce polymer components directly from a computer representation of the part. Though SLA is being recognized as an innovative technology, it still cannot be used to fully practical application since it lacks of dimensional accuracy compared to conventional process. If the shrinkage were perfectly uniform and no distortion took place, excellent part accuracy could still be achieved through and appropriate scaling factor when generating the build file. However, in certain geometries involving intersecting thick and thin sections, nonuniform retrain shrinkage becomes the engine of part distortion. In order to improve the part accuracy of SLA, this paper evaluates how largely each parameter of SLA contributes to the part accuracy and estimates the optimal set of parameter which minimizes the dimension error of the test part, "Slab (100mm$\times$100mm$\times$2mm)"and "scale bar"part. Three control parameters such as critical exposure, generation depth and fill cure depth are used.