• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.10a
• /
• pp.467-471
• /
• 1997

AGV is very proper equipment for Port Automation. AGV must have Obstacle Detection System(ODS) for port automation. Obstacle Detection System must have some functions. It must be able to classify some specified object from background data. And it must be able to track classified objects. Finally, ODS must determine its next action for safe cruise whether it must do emergency stop or it must speed down or it must change it track. For these functions, ODS can have many different structure. In this paper, we will propose one structure among some possible own which is under construction.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.6
• /
• pp.477-484
• /
• 2016

Small industrial appliances such as fan, audio, electric rice cooker mostly consist of ABS, PP, PS materials. In colored plastics, it is possible to classify by near infrared(NIR) spectroscopy, while in black plastics, it is very difficult to classify black plastic because of the characteristic of black material that absorbs the light. So the RBFNNs pattern classifier is introduced for sorting electrical and electronic waste plastics through LIBS(Laser Induced Breakdown Spectroscopy) spectrometer. At the preprocessing part, PCA(Principle Component Analysis), as a kind of dimension reduction algorithms, is used to improve processing speed as well as to extract the effective data characteristics. In the condition part, FCM(Fuzzy C-Means) clustering is exploited. In the conclusion part, the coefficients of linear function of being polynomial type are used as connection weights. PSO and 5-fold cross validation are used to improve the reliability of performance as well as to enhance classification rate. The performance of the proposed classifier is described based on both optimization and no optimization.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.4
• /
• pp.569-577
• /
• 2018

In this paper, with the aid of information which is included within data, preprocessing algorithm-based black plastic classifier is designed. The slope and area of spectrum obtained by using laser induced breakdown spectroscopy(LIBS) are analyzed for each material and its ensuing information is applied as the input data of the proposed classifier. The slope is represented by the rate of change of wavelength and intensity. Also, the area is calculated by the wavelength of the spectrum peak where the material property of chemical elements such as carbon and hydrogen appears. Using informations such as slope and area, input data of the proposed classifier is constructed. In the preprocessing part of the classifier, Principal Component Analysis(PCA) and fuzzy transform are used for dimensional reduction from high dimensional input variables to low dimensional input variables. Characteristic analysis of the materials as well as the processing speed of the classifier is improved. In the condition part, FCM clustering is applied and linear function is used as connection weight in the conclusion part. By means of Particle Swarm Optimization(PSO), parameters such as the number of clusters, fuzzification coefficient and the number of input variables are optimized. To demonstrate the superiority of classification performance, classification rate is compared by using WEKA 3.8 data mining software which contains various classifiers such as Naivebayes, SVM and Multilayer perceptron.

• Current Optics and Photonics
• /
• v.3 no.2
• /
• pp.150-153
• /
• 2019

Larynx cancer is a potentially terminal and severe type of neck and head cancer in which malignant cells start to grow and spread upwards in the larynx, or voice box. Smoking tobacco, drinking hot beverages and drinking alcohol are the main risk factors for these tumors. In this study, we aimed to develop a precise, accurate and rapid chemometrics assisted Raman spectroscopy method for diagnosis of larynx cancer in deparaffinized tissue samples. In the proposed method, samples were deparaffinized and 20 microns of each tissue were located on a coverslip. Both healthy (n = 13) and cancerous tissues (n = 13) were exposed to a Raman laser (785 nm) and excitations were recorded between wavenumbers of $50{\sim}1500cm^{-1}$. An Orthogonal Partial Least Square algorithm was applied to evaluate the Raman spectrum obtained. Sensitivity and specificity of the proposed method is high enough with the aid of Principal Component Analysis (PCA) to test the whole model. Healthy and cancerous tissues were accurately and precisely clustered. A rapid, easy and precise diagnosis algorithm was developed for larynx cancer. By this method, some useful data about differences in biomolecules of each group (phospholipids, amides, tyrosine, phenylalanine collagen etc.) was also obtained from the spectra. It is claimed that the optimized method has a great potential for clustering and separating tumor tissues from healthy ones. This novel, rapid, precise and objective diagnosis method may be an alternative for the conventional methods in literature for diagnosis of larynx cancer.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1034-1039
• /
• 2005

SALM(Simultaneous localization and mapping) and AI(Artificial intelligence) have been active research areas in robotics for two decades. In particular, localization is one of the most important tasks in mobile robot research. Until now expensive sensors such as a laser sensor have been used for mobile robot localization. Currently, the proliferation of RFID technology is advancing rapidly, while RFID reader devices, antennas and tags are becoming increasingly smaller and cheaper. So, in this paper, the smart floor using passive RFID tags is proposed and, passive RFID tags are mainly used for identifying location of the mobile robot in the smart floor. We discuss a number of challenges related to this approach, such as tag distribution (density and structure), typing and clustering. In the smart floor using RFID tags, the localization error results from the sensing area of the RFID reader, because the reader just knows whether the tag is in the sensing range of the sensor and, until now, there is no study to estimate the heading of mobile robot using RFID tags. So, in this paper, two algorithms are suggested to. The Markov localization method is used to reduce the location(X,Y) error and the Kalman Filter method is used to estimate the heading($\theta$) of mobile robot. The algorithms which are based on Markov localization require high computing power, so we suggest fast Markov localization algorithm. Finally we applied these algorithms our personal robot CMR-P3. And we show the possibility of our probability approach using the cheap sensors such as odometers and RFID tags for mobile robot localization in the smart floor

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1797-1801
• /
• 2005

SLAM(Simultaneous localization and mapping) and AI(Artificial intelligence) have been active research areas in robotics for two decades. In particular, localization is one of the most important issues in mobile robot research. Until now expensive sensors like a laser sensor have been used for the mobile robot's localization. Currently, as the RFID reader devices like antennas and RFID tags become increasingly smaller and cheaper, the proliferation of RFID technology is advancing rapidly. So, in this paper, the smart floor using passive RFID tags is proposed and, passive RFID tags are mainly used to identify the mobile robot's location on the smart floor. We discuss a number of challenges related to this approach, such as RFID tag distribution (density and structure), typing and clustering. In the smart floor using RFID tags, because the reader just can senses whether a RFID tag is in its sensing area, the localization error occurs as much as the sensing area of the RFID reader. And, until now, there is no study to estimate the pose of mobile robot using RFID tags. So, in this paper, two algorithms are suggested to. We use the Markov localization algorithm to reduce the location(X,Y) error and the Kalman Filter algorithm to estimate the pose(q) of a mobile robot. We applied these algorithms in our experiment with our personal robot CMR-P3. And we show the possibility of our probability approach using the cheap sensors like odometers and RFID tags for the mobile robot's localization on the smart floor.

• Tunnel and Underground Space
• /
• v.26 no.1
• /
• pp.46-58
• /
• 2016

In this study, a LIDAR laser scanner was used to scan a rock slope around Mt. Gwanak and to produce point cloud from which directional information of rock joint surfaces shall be extracted. It was analyzed using two different algorithms, i.e. Ball Pivoting and Wrap algorithm, and four sampling intervals, i.e. raw, 2, 5, and 10 cm. The results of Fuzzy K-mean clustering were analyzed on the stereonet. As a result, the Ball Pivoting and Wrap algorithms were considered suitable for extraction of rock surface orientation. In the case of 5 cm sampling interval, both triangulation algorithms extracted the most number of the patch and patched area.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.1
• /
• pp.47-53
• /
• 2006

Localization is a essential technology for mobile robot to work well. Until now expensive sensors such as laser sensors have been used for mobile robot localization. We suggest RFID tag based localization system. RFID tag devices, antennas and tags are cheap and will be cheaper in the future. The RFID tag system is one of the most important elements in the ubiquitous system and RFID tag will be attached to all sorts of goods. Then, we can use this tags for mobile robot localization without additional costs. So, in this paper, the smart floor using passive RFID tags is proposed and, passive RFID tags are mainly used for identifying mobile robot's location and pose in the smart floor. We discuss a number of challenges related to this approach, such as tag distribution (density and structure), typing and clustering. When a mobile robot localizes in this smart floor, the localization error mainly results from the sensing range of the RFID reader, because the reader just ran know whether a tag is in the sensing range of the sensor. So, in this paper, we suggest two algorithms to reduce this error. We apply the particle filter based Monte Carlo localization algorithm to reduce the localization error. And with simulations and experiments, we show the possibility of our particle filter based Monte Carlo localization in the RFID tag based localization system.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.59 no.3
• /
• pp.252-262
• /
• 2014

Sorghum seed is traditionally used as secondary food sources in addition to rice in Korea. While the hypoglycemia regulating phytochemicals have been found in sorghum seed, peptides related with hypoglycemia never been studied before. To obtain the peptide characteristics and the specifically high-expressed peptides in hypoglycemic sorghum seed, peptide profiles of seven hypoglycemic and five non-hypoglycemic sorghum lines bred in RDA were determined using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS). The twelve sorghum lines exhibited 104 peptides on CM10 protein chip array (weak cation exchange) and 95 peptides on Q10 (weak cation exchange) in the molecular mass range from 2,000 to 20,000 Da. Heat map via supervised hierarchical clustering of the significantly different peptides (p < 0.01) in peak intensity among the 12 lines effectively revealed the specifically upregulated peptides in each line and distinguished between 7 hypoglycemic and 5 non-hypoglycemic lines. Through the comparison with hypoglycemic and non-hypoglycemic lines, 10 peptides including 2231.6, 2845.4, 2907.9, 3063.5, 3132.6, 3520.8, 4078.8, 5066.2, 5296.5, 5375.5 Da were specifically high-expressed in hypoglycemic lines at p < 0.00001. This study characterized seed peptides of 12 sorghums and found ten peptides highly expressed for hypoglycemic sorghum lines, which could be used as peptide biomarkers for identification of hypoglycemic sorghum.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2000.02a
• /
• pp.113-113
• /
• 2000

직접천이형 wide band gap(3.4eV) 반도체중의 하나인 GaN를 청색 및 자외선 laser diode, 고출력 전자장비 등으로 응용하기 위해서는 낮은 접합저항을 갖는 Ohmic contact이 선행되어야 한다. 그러나 만족할만한 p-type GaN의 Ohmic contact은 아직 실현되고 있지 못하며, 이는 GaN와 접합 금속과의 구체적인 반응의 연구를 필요로 한다. 본 연구에서 앞서 Pt, Pt, Ni등의 late transition metal을 p-GaN에 접합시킨 결과 이들은 접합 당시 비교적 평탄하나 후열 처리과정에서 비교적 낮은 온도에서 기판과 열팽창계수의 차이로 인하여 평탄성을 잃어버리면서 barrier height가 증가한다는 사실을 확인하였다. 따라서 본 연구에서는 이러한 열적 불안정성을 극복하기 위하여 Ni과 Pd를 차례로 증착하고 가열하면서 interfacial reaction, film morphology, Fermi level의 움직임을 monchromatic XPS(x-ray photoelectron spectroscopy) 와 SAM(scanning Auger microscopy) 그리고 ex-situ AFM을 이용하여 밝히고자 하였다. 특히 후열처리에 의한 계면 반응에 수반되는 구성 금속원소 간의 합금현상과 금속 층의 평탄성이 밀접한 관계가 있다는 것을 확인하였다. 이러한 합금과정에서 나타나는 금속원소들의 중심 준위의 이동을 체계적으로 규명하기 위해서 Pd1-xNix와 Pd1-xGax 합금들의 표준시료를 arc melting method로 만들어 농도에 따른 금속원소들의 중심 준위의 이동을 측정하여, Pd/Ni/p-GaN 및 Ni/Pd/p-GaN 계에서 열처리 온도에 따른 interfacial reaction을 확인하였다. 그 결과 두 계가 상온에서 nitride 및 alloy를 형성하지 않고 고르게 증착되고, 열처리 온도를 40$0^{\circ}C$에서 $650^{\circ}C$까지 증가시킴에 따라 계면반응의 부산물인 metallic Ga은 증가하고 있으마 nitride는 여전히 형성되지 않는 것을 확인하였다. 증착당시 Ni이 계면에 있는 Pd/Ni/p-GaN의 경우에는 52$0^{\circ}C$까지의 열처리에 의하여 Ni과 Pd가 골고루 섞이고 그 평탄성도 유지되고 barier height의 변화도 없었다. 더 높은 $650^{\circ}C$ 가열에 의해서는 surface free energy가 작은 Ga의 활발한 편석 현상으로 인해 표면은 Ga이 풍부한 Pd-Ga의 합금층으로 덮이고, 동시에 작은 pinhole들이 발생하며 barrier height도 0.3eV 가량 증가하게 된다. 반면에 증착당시 Pd이 계면에 있는 Ni/Pd/p-GaN의 경우에는 40$0^{\circ}C$의 가열까지는 두 금속이 그들 계면에서부터 섞이나, 52$0^{\circ}C$의 가열에 의해 이미 barrier height가 0.2eV 가량 증가하기 시작하였다. 더 높은 $650^{\circ}C$가열에 의해서는 커다란 pinhole, 0.5eV 가량의 barrier height 증가, Pd clustering이 동시에 관찰되었다. 따라서 Ni과 Pd의 일함수는 물론 thermal expansion coefficient가 거의 같으며 surface free energy도 거의 일치한다는 점을 감안하면, 이렇게 뚜렷한 열적 안정성의 차이는 GaN와 contact metal과의 반응시작 온도(disruption onset temperature)의 차이에 기인함을 알 수 있었다. 즉 계면에서의 반응에 의해 편석되는 Ga에 의해 박막의 strain이 이완되면, pinhole 등의 박막결함이 줄어 들고, 이는 계면의 N의 out-diffusion을 방지하여 p-type GaN의 barrier height 증가를 막게 된다.