• Journal of the Korean Institute of Intelligent Systems
• /
• v.7 no.3
• /
• pp.65-71
• /
• 1997

With the emergence of neural network, there is a revived interest in identification of nonlinear systems. So in this paper, to identify unknown nonlinear systems dynamically we propose DPNN(Dynamic Polynomial Neural Network) using GMDH (Group Method of Data Handling) algorithm. The dynamic system identification using GMDH consists of applying a set of inputloutput data to train the network by dynamically computing the necessary coeffici1:nt sets. Then, MRAC(Mode1 Reference Adaptive Control) is designed to control nonlinear systems using DPNN. In the result, we can see that the modelling and control using DPNN work well by computer simulation.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.4
• /
• pp.33-38
• /
• 2021

This paper presents a comparative study between the parameter-optimized Pacejka model and artificial neural network model for the tire force estimation. The two different approaches are investigated and compared in this study. First, offline optimization is conducted based on Pacejka Magic Formula model to determine the proper parameter set for the minimization of tire force error between the model and test data set. Second, deep neural network model is used to fit the model to the tire test data set. The actual tire forces are measured using MTS Flat-Track test platform and the measurements are used as the reference tire data set. The focus of this study is on the applicability of machine learning technique to tire force estimation. It is shown via the regression results that the deep neural network model is more effective in describing the tire force than the parameter-optimized Pacejka model.

• Journal of Positioning, Navigation, and Timing
• /
• v.13 no.1
• /
• pp.103-110
• /
• 2024

In this paper, a test bed for real-time network Real-Time Kinematic (RTK) research was constructed using reference stations of the NGII. A group of candidate station networks was derived, including three stations in Seoul. The group consisted of four stations with a distance of less than 100 km between them. Among several candidates, a network composed of stations with short distances between them and demonstrating good data quality for all reference stations was selected as the test bed. After collecting real-time data in Radio Technical Committee for Maritime services (RTCM) format from the selected stations and conducting a noise analysis on measurements, mm-level carrier phase measurement noise was confirmed. Afterwards, the user set the reference station inside the test bed and analyzed the network RTK positioning performance of the MAC method using the GPS L1 frequency as post-processing. From the result of the analysis it was confirmed that the residual error for all users was within 10 cm after applying the correction. Additionally, after determining integer ambiguities through Least-squares AMBiguity Decorrelation Adjustment (LAMBDA), it was confirmed that the fix rate was 100%, and all ambiguities were resolved as true values.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.2 s.308
• /
• pp.33-39
• /
• 2006

This paper presents the modified LVQ method for performance improvement of pattern classification. The proposed method uses the skewness of probability distribution between the input vectors and the reference vectors. During training, the reference vectors are closest to the input vectors using the probabilistic distribution of the input vectors, and they are positioned to approximate the decision surfaces of the theoretical Bayes classifier. In order to verify the effectiveness of the proposed method, we performed experiments on the Gaussian distribution data set, and the Fisher's IRIS data set. The experimental results show that the proposed method considerably improves on the performance of the LVQ1, LVQ2, and GLVQ.

• Journal of Korean Society of Archives and Records Management
• /
• v.20 no.4
• /
• pp.41-58
• /
• 2020

Although numerous studies have noted the importance of data sets in government information systems, the practical management of data sets has yet to be developed. Under these circumstances, the National Archives of Korea designated data set management as a major project in 2020, initiating full-scale management work. Despite these efforts, the records center, which will conduct management, expressed great concern for the new project. As such, this study identifies problems in managing data sets and searches for possible improvements through a comparison with existing public data projects by public institutions. In particular, the following materials were analyzed: laws, notices, guidelines, and publications issued by the ministries. Based on the results, several measures were proposed as part of an improvement plan for data set management: (1) the utilization of government functional classification as a reference, (2) the reorganization of the table, and (3) data linkage with related systems.

• Journal of Korean Society of Forest Science
• /
• v.111 no.4
• /
• pp.603-612
• /
• 2022

The study aims were to evaluate a machine-learning, algorithm-based, forest biomass-estimation model to estimate subnational forest biomass and to comparatively analyze REDD+ forest reference emission levels. Time-series Landsat satellite imagery and ESA Biomass Climate Change Initiative information were used to build a machine-learning-based biomass estimation model. The k-nearest neighbors algorithm (kNN), which is a non-parametric learning model, and the tree-based random forest (RF) model were applied to the machine-learning algorithm, and the estimated biomasses were compared with the forest reference emission levels (FREL) data, which was provided by the Paraguayan government. The root mean square error (RMSE), which was the optimum parameter of the kNN model, was 35.9, and the RMSE of the RF model was lower at 34.41, showing that the RF model was superior. As a result of separately using the FREL, kNN, and RF methods to set the reference emission levels, the gradient was set to approximately -33,000 tons, -253,000 tons, and -92,000 tons, respectively. These results showed that the machine learning-based estimation model was more suitable than the existing methods for setting reference emission levels.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1618-1622
• /
• 2004

In this paper, the linearly interpolated PRC(Pseudo Range Correction) regenerating algorithm was applied to improve the DGPS(Differential Global Positioning System) positioning accuracy at user's spot by using the various PRC information obtained from multi-DGPS reference stations. The PRC information of each GPS satellite is not varying rapidly; it is possible to assume that the variation of PRC information of each GPS satellite is linear. So the linearly interpolated PRC regenerating algorithm can be applied to improve the DGPS positioning accuracy at user's spot by using the various PRC information obtained from multi-DGPS reference stations. To test the performance of the linearly interpolated PRC regenerating algorithm, maritime DGPS reference stations' PRC data was used in RTCM format. 11 maritime DGPS reference stations are in service providing DGPS information to public since 1999. Two set of 3 DGPS reference stations are selected to compare the performance of the linearly interpolated PRC regenerating algorithm. The DGPS positioning accuracy was dramatically improved about 40%. Linearly interpolated PRC regenerating algorithm adopted multi-channel DGPS receiver will be developed in near future.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.393-402
• /
• 2013

The Mahalanobis-Taguchi System (MTS) is a diagnostic and predictive method for multivariate data. In the MTS, the Mahalanobis space (MS) of reference group is obtained using the standardized variables of normal data. The Mahalanobis space can be used for multi-class classification. Once this MS is established, the useful set of variables is identified to assist in the model analysis or diagnosis using orthogonal arrays and signal-to-noise ratios. And other several techniques have already been used for classification, such as linear discriminant analysis and logistic regression, decision trees, neural networks, etc. The goal of this case study is to compare the ability of the Mahalanobis-Taguchi System and logistic regression using a data set.

• Proceedings of the KSR Conference
• /
• 2006.11b
• /
• pp.322-327
• /
• 2006

Banwol bridge with steel plate girder and Pyongtaek bridge with PSC bos girder have been operated maintenance measuring system by the Seoul-Chonan of Kyongbu express railway. By analyzing the theoretical and experimental values of design load for these two bridge, the establishment of reference maintenance for measuring items was deduced from research. Two materials, steel and concrete plates, were considered as the upper structure. Actual measurement data for the behavior under speed, structural analysis results, and the presented references were analyzed and used to set up the reference establishment. The measuring items are stress(strain), displacement, dynamic acceleration, expansion movement, and dynamic frequency. The maintenance reference was established by comparing analytical and measuring values of the five items with respect to structural state class.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.10
• /
• pp.1338-1346
• /
• 1995

In this paper, we suggest the implementation of a DGPS system using two low cost commercial C/A code GPS modules and modems and its efficient operational techniques to provide DGPS service which guarantees the position accuracy of better than 10 meters for more users. The proposed DGPS system can be implemented easil at low cost because it needs a GPS module and a modem for each reference station and user. The reference station makes plans of the receiving schedule from the satellite set at each period and then provides the correction data for various satellite sets in a period. The main contribution of this paper is that users can utilize the correction data continuously and efficiently through the recursive least squares lattice filtering method. Experimental results show the position accuracy of better than 10 meters using the suggested DGPS system in almost real time.