In this study, we present an algorithm for indoor robot position estimation. Estimating the position of an indoor robot using a fixed imaging device obviates the need for complex sensors or hardware, enabling easy estimation of absolute position through marker recognition. However, location estimation becomes impossible when the device moves away from the surrounding obstacles or the screen, presenting a significant drawback. To solve this problem, we propose an algorithm that improves the precision of robot indoor location estimation using a Gaussian Mixture Model(GMM) and a Kalman filter estimation model. We conducted an actual robot operation experiment and confirmed accurate position estimation, even when the robot was out of the image.

This study explored the role of written languaging (WL) in response to automated written corrective feedback (AWCF) in L2 accuracy improvement in English classrooms at a university in China. A total of 254 freshmen enrolled in intermediate composition classes participated, and they wrote 4 essays and received AWCF. A half of them engaged in WL after receiving correction, while the other half did not. To measure whether WL could contribute to L2 accuracy improvement, error correction tests were conducted three times. AWCF targeted all types of errors, but the study focused on investigating six language features (nouns, conjunctions, subject-verb agreement, sentence constituents, articles, and prepositions) known as the most frequent errors intermediate students make. The results proved that WL had a positive effect on accuracy improvement overall. However, when individual language features were considered separately, the effects were not the same. The inherent complexity of the features along with L1 influence appeared to affect the effects of WL for accuracy improvement.

At ship design stage, the maneuverability is generally estimated based on the empirical formula or the computational fluid dynamic (CFD), which is one of the numerical simulation methods. Using the hydrodynamic derivatives derived through these methods can quantitatively estimate the maneuverability of target vessels and evaluate indirect maneuverability. Nevertheless, research on estimating maneuverability is insufficient for ships not subject to IMO maneuverability standard, especially fishing vessels, and even at the design stage, the empirical formula developed for merchant ships is applied without modification. An estimation error may occur due to the empirical formula derived from the regression analysis results of a model test if the empirical formula developed for merchant ships with different hull shapes is applied to fishing vessels without any modification. In this study, the modified empirical formula that can more accurately estimate the fishing vessel's maneuverability was derived by including the hull shape parameter of target fishing trawlers in the regression analysis process that derives Kijima et al. (1990) formula. As a result, the modified empirical formula showed an average estimation error of 6%, and the result improved the average error of 49% of Kijima et al. (1990) formula developed for merchant ships.

본 연구에서는 점근해석 및 논로컬 이론에서 요구하는 4차 이상의 고차 미분근사를 수행하기 위하여 계방정식에 혼합변분이론을 적용하여 MLS 차분법으로부터 구해지는 고차 미분근사의 정확도를 큰 폭으로 향상시킨다. 또한, MLS 차분법에 존재하는 세 가지 조건변수에 따른 고차미분근사의 정확도를 비교·분석한다. 혼합변분이론의 합응력을 후처리하여 변위의 미분을 근사할 경우 기존의 변위장 기반 계방정식의 차분 결과에 비해 미분 차수가 2차 낮아진다. 해석 범위내 절점 수가 과도하게 많거나 기저 차수가 클 경우 MLS 차분법의 영향영역 내에서 과적합(overfitting)이 발생한다. 또한 영향영역이 최적 범위 이상으로 넓어질 경우 근사의 정확도가 떨어진다. 위 내용을 사인 하중을 받는 단순지지보 수치예제로부터 확인하였다.

PURPOSES : The purpose of this study is to compare applicability, explanation power, and flexibility of traffic accident models between estimating model using the statistical method and the machine learning method. METHODS: In order to compare and analyze traffic accident models between model estimated using the statistical method and machine learning method, data acquisition was conducted, and traffic accident models were estimated using statistical methods such as negative binomial regression model, and machine learning methods such as a generalized regression neural network (GRNN). Then, the fitness of model as R2, root mean square error (RMSE), mean absolute percentage error (MAPE), accuracy, etc., were determined to compare the traffic accident models. RESULTS: The results showed that the annual average daily traffic (AADT), speed limits, number of lanes, land usage, exclusive right turn lanes, and front signals were significant for both traffic accident models. The GRNN model of total traffic accidents had been better statistical significant with R2: 0.829, RMSE: 2.495, MAPE: 32.158, and Accuracy: 66.761 compared with the negative binomial regression model with R2: 0.363, RMSE: 9.033, MAPE: 68.987, and Accuracy: 8.807. The GRNN model of injury traffic accidents also showed similar results of model’s statistical significance. CONCLUSIONS: Traffic accident models estimated with GRNN had better statistical significance compared with models estimated with statistical methods such as negative binomial regression model.

In this work, measured data from thickness gauge behind of the finishing mill stand was used for increasing accuracy of thickness control in plate rolling process. The Automatic Gauge Control (AGC) system of the mill could control roll gap for the remaining rolling passes based on the difference between measured and calculated thickness. This work was possible with some modification in software product system to use measured data without additional installation of equipment. As a result, the accuracy of thickness has been increased up to 31%. The accuracy of thickness control was defined as a standard deviation of the differences between target and measured final thickness of average.

The aim of this paper is to describe a strategy to provide improved accuracy of primary navigation parameters to the Indian regional Navigational satellites. The broadcast parameters of the Navic satellites can be classified under two category, one is the validity period of 7200s called long period normal sets and second one is the validity period of 900s called short period contingency sets of primary navigation parameters. The Navic satellite constellation works with minimum number of synchronous satellites to provide the best user solution within India and boundaries. Thus the outage of any one or more satellites will increase the Dilution of Precision (DOP) and degrades the user solution for any sudden anomalous behaviour of the measurements. This anomalous behaviour (called events) arises due to on-board frequency variations, frequent planned Station Keeping (SK) operations because of synchronous orbits, IRNSS system time scale switch-over or the combination of events. In loop back the broadcast parameters accuracy was continuously monitored through Line of Sight (LOS) range error from the observed IRNSS ground reference stations range measurements. In begin conditions the Least square (Lsq) based solutions were accurate and the observed LOS error were as expected. But post the occurrence of any event, the accumulation of batch data starts freshly for least square solution. The obtained Lsq based solution was inaccurate due to lack of data under many circumstances. During the events occurrence to minimize the outage duration of the satellite contingency sets were generated using continuously running Kalman Filter based near real time estimation using one-way measurements. But the filter estimated state may not be optimum at the particular epoch. Hence a strategy of forward and backward approach (FBF) was adopted just before parameter uplink through sequential (KF) for adaptation of real behaviours of the measurements to providing an improved optimal solution.

We used an etching process to control the line-width of screen printed Ag paste patterns. Ag paste was printed on anodized Al substrate to produce a high power LED. In general, Ag paste spreads or diffuses on anodized Al substrate in the process of screen printing; therefore, the line-width of the printed Ag paste pattern increases in contrast with the ideal line-width of the pattern. Smudges of Ag paste on anodized Al substrate were removed by neutral etching process without surface damage of the anodized Al substrate. Accordingly, the line-width of the printed Ag paste pattern was controlled as close as possible to the ideal line-width. When the etched Ag paste pattern was used as a seed layer for electroless Ni plating, the line width of the plated Ni film was similar to the line-width of the etched Ag paste pattern. Finally, in pattern formation by Ag paste screen printing, we found that the accuracy of the line-width of the pattern can be effectively improved by using an etching process before electroless Ni plating.

Isosteric heat of hydrogen adsorption is one of the most important parameters required to describe solid-state hydrogen storage systems. Typically, it is calculated from adsorption isotherms measured at 77K (liquid N2) and 87K (liquid Ar). This simple calculation, however, results in a high degree of uncertainty due to the small temperature range. Therefore, the original Sievert type setup is upgraded using a heating and cooling device to regulate the wide sample temperature. This upgraded setup allows a wide temperature range for isotherms (77K ~ 117K) providing a minimized uncertainty (error) of measurement for adsorption enthalpy calculation and yielding reliable results. To this end, we measure the isosteric heats of hydrogen adsorption of two prototypical samples: activated carbon and metal-organic frameworks (e.g. MIL-53), and compared the small temperature range (77~87K) to the wide one (77K ~ 117K).

풍력 자원의 단기 예측 가능성은 풍력 발전 단지의 경제적 타당성을 평가하는 중요한 요소이다. 본 연구에서는 풍력 자원의 단기 예측 가능성을 향상시키는 방법의 하나로 베이지안 칼만 필터를 후처리 과정으로 적용하였다. 이때 추정된 모델과 관측 데이터의 상관관계를 평가하기 위하여 일정 시간 동안 베이지안 칼만 훈련 기간이 요구된다. 본 연구는 여러 훈련 기간에 따라 예측 특성을 정량적으로 분석하였다. 태백 지역에서는 3일 단기 베이지안 칼만 훈련으 로 기온과 풍속을 예측하는 것이 다른 훈련 기간을 적용할 때보다 우수한 예측 성능을 보였다. 반면 이어도는 6일 이 상의 베이지안 칼만 필터의 훈련 기간을 적용한 경우 가장 좋은 예측 성능을 나타낸다. WRF 예측 성능이 떨어지는 사 례에서 베이지안 칼만 필터의 예측 성능향상이 뚜렷하게 나타나며, 반대로 WRF 예측이 정확한 지점에서는 필터적용에 따른 성능향상 정도가 약한 경향을 가진다.

PURPOSES : The purpose of this study is to develop a method for improving the accuracy of analysis results obtained from a twodimensional (2-D) numerical analysis model of continuously reinforced concrete pavement (CRCP). METHODS: The analysis results from the 2-D numerical model of CRCP are compared with those from more rigorous three-dimensional (3- D) models of CRCP, and the relationships between the results are recognized. In addition, the numerical analysis results are compared with the results obtained from field experiments. By performing these comparisons, the calibration factors used for the 2-D CRCP model are determined. RESULTS : The results from the comparisons between 2-D and 3-D CRCP analyses show that with the 2-D CRCP model, concrete stresses can be overestimated significantly, and crack widths can either be underestimated or overestimated by a slight margin depending on the assumption of plane stress or plane strain. The behaviors of crack width in field measurements are comparable to those obtained from the numerical model of CRCP. CONCLUSIONS: The accuracy of analysis results from the 2-D CRCP model can be improved significantly by applying calibration factors obtained from comparisons with 3-D analyses and field experiments.

In this paper, the author devised new decision recommendation ordering method of items attributed by age to improve accuracy of recommender system. In conventional recommendation system, recommendation order is decided by high order of preference prediction. However, in this paper, recommendation accuracy is improved by decision recommendation order method that reflect age attribute of target customer and neighborhood in preference prediction. By applying decision recommendation order method to recommender system, recommendation accuracy is improved more than conventional ordering method of recommendation.

In this paper, the author proposed following two methods to improve the accuracy of the recommender system. First, in order to classify the users more accurately, the author used a EMC(Expanded Moving Center) heuristic algorithm which improved clustering accuracy. Second, the author proposed the Neighborhood-oriented preference prediction method that improved the conventional preference prediction methods, so the accuracy of the recommender system is improved. The test result of the recommender system which adapted the above two methods suggested in this paper was improved the accuracy than the conventional recommendation methods.

We present a method to improve the RV (radial velocity) measurements accuracy by using telluric lines. Telluric lines are used to estimate the wavelength scale drift over the detector of the spectrograph. In the case of BOES, the Echelle spectrograph at BOAO (Bohyunsan Optical Astronomical Observatory), the wavelength scale drift can be several hundreds m/s over 24 hours. Due to the wavelength scale drift, the RV measurements accuracy of BOES is limited to several hundreds m/s. By estimating the wavelength scale drift by telluric lines, we can remove its effect to improve the RV measurements accuracy to about 40 m/s.

주소는 현대사회에서 사람들이 위치를 개념화하는 기본적인 수단 중 하나이며, 위치의 정보를 지니는 주소는 개인행태나 특정 사상의 공간적 분포 분석 시 사용되고 있다. 현재 우리나라의 주소체계는 지번주소와 도로명주소가 병행되어 사용되고 있다. 우리나라의 주소표시방법인 지번주소체계는 산업화⋅도시화의 진전으로 도시의 구조가 복잡해지고 거대화됨에 따라 주소의 위치 표시기능에 대해 문제점이 많이 지적되었기 때문이다. 본 연구에서는 도로명주소체계는 도로망기반의 지오코딩이 가능해짐에 따라 위치정확도 를 평가하고 이를 개선시킬 수 있는 방안들을 제시하였다. 실험결과 도로세그먼트를 수정하지 않은 상태에서 지오코딩의 위치오차는 평균 6.88m이었으나, 도로세그먼트 수정을 통해 평균 3.94m로 위치정확도가 향상되었다. 이러한 결과는 향후 주소자료의 GIS데이터베이스 구축에 효과적으로 이용될 수 있다.

This study investigates the effects of indirect and direct error feedback on Korean college students’ accuracy improvement in writing and their responses to the feedback they received. The 32 participants of the study were divided into four groups of different error feedback conditions: indirect feedback, indirect feedback along with metalinguistic feedback, direct feedback, and direct feedback along with metalinguistic feedback. Direct feedback was full, explicit error correction, while indirect feedback was coded error correction in which errors were underlined and marked with code. Metalinguistic feedback was marginal explanations about errors. Analysis of error rates in the students’ first and last essay writing did not show statistically significant difference in the accuracy improvement of the four groups. In other words, indirect error feedback was not more effective than direct feedback. However, the gain score in each group showed that indirect feedback and direct feedback were more effective when they were combined with metalinguistic feedback. In particular, the students’ learning diary clearly indicated that the students valued error feedback on their writing and that follow-up feedback was crucial in providing indirect error feedback.

기선장에 따른 GPS 위치 정도 및 분포를 측정하기 위하여 2002년 5월 29일부터 6월 2일까지 2대의 GPS수신기 (L1, 12채널)를 이용하여 고정점에서 30분에서 24시간동안 관측하는 실험을 수행하였다. 기준국(PKNU)과 이동, 국립지리원의 GPS 상시관측소에서 수신한 GPS 데이터는 동적 및 정적 후처리방식으로 처리하였으며, 그 결과를 요약하면 다음과 같다. 1. 2002년 5월 31일 차단각 (cut-off angle) 15˚에서 6시간 이상 관측 가능한 위성은 16개였고, 위성은 주로 동서방향으로 배치되어 있었다. 10분마다 평균한위성 수와 GDOP는 각각 8개, 3.89이었다. 2. 후처리 전후의 GPS 평균위치는 모두 기준점에서 남서쪽으로 편위 (단독 측위 : 1.17m, 후처리 : 평균 0.43m)되었다. 확율원 오차는 단독 측위가 6.65m이었고 이를 동적 후처리하면 단독 측위의 약 33.8%(표준편차 0=17.2), 정적 후처리하면 단독 측위의 약 5.3% (0=2.2)로 감소시 킬 수 있었다. 3. 정적 후처리방식으로 구한 기선장 x(Km)와 확률원 오차 y(m) 의 관계식은 y = 0.0016x + 0006 (R2=0.87)이었다. 본 연구에서 사용한 1주파용 GPS 수신기를 이용하여 정적 후처리방식으로 위치를 측정하는 경우, 기선장을 100Km 이내로 하고 GPS 수신시간을 30분 이상으로 하면 확율원 오차 20 Cm 이내의 측위가 가능함을 알 수 있었다.

Underwater acoustic positioning systems have been extensively used not only in surface position fixing but also in underwater position fixing. Recently, these systems have been applied in the field of installation and underwater inspection of offshore platforms etc. But in these systems are included the fixing errors as results of a signal with additive noise and irregular motion of vessel by ocean waves. To improve the accuracy of the position fixing a Kalman filter is applied to the supershort baseline (SSBL) acoustic positioning system with beacon mode in noise conditions. The position data obtained by the Kalman filter is compared with raw position data and it is confirmed in the simulation that the former is more accurate than the latter. And an indicator monitoring the filtering effect is described while ship's moving.