• Journal of the Korea Society of Computer and Information
• /
• v.14 no.11
• /
• pp.159-167
• /
• 2009

The vehicles play a significant role in modern people's life as economy grows. The development of car navigation system(CNS) provides various convenience because it shows the driver where they are and how to get to the destination from the point of source. However, the existing map-based CNS does not consider any environments such as traffic congestion. Given the same starting point and destination, the system always provides the same route and the required time. This paper proposes a path planning method with traffic prediction by applying historical driving information to the Fuzzy theory and Bayesian update. Fuzzy theory classifies the historical driving information into groups of leaving time and speed rate, and the traffic condition of each time zone is calculated by Bayesian update. An ellipse area including starting and destination points is restricted in order to reduce the calculation time. The accuracy and practicality of the proposed scheme are verified by several experiments and comparisons with real navigation.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.991-997
• /
• 2020

Reinforcement learning has been applied mainly in sequential decision-making problems. Especially in recent years, reinforcement learning combined with neural networks has brought successful results in previously unsolved fields. However, reinforcement learning using deep neural networks has the disadvantage that it is too complex for immediate use in the field. In this paper, we implemented path planning algorithm for mobile robots using Q-learning, one of the easy-to-learn reinforcement learning algorithms. We used real-time Q-learning to update the Q-table in real-time since the Q-learning method of generating Q-tables in advance has obvious limitations. By adjusting the exploration strategy, we were able to obtain the learning speed required for real-time Q-learning. Finally, we compared the performance of real-time Q-learning and DQN.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.705-708
• /
• 2005

Mobile Mapping Systems using the vehicle equipped the GPS, IMU, CCD Cameras is the effective system for the management of the road facilities, update of the digital map, and etc. They must provide users with the sensor data which is acquired by Mobile Mapping Systems in real-time so that users can process what they want by using the latest data. But it' s not an easy process because the amount of sensor data is very large, particularly image data to be transmitted. So it is necessary to reduce the amount of image data so that it is transmitted effectively. In this study, the effective method was suggested for the compression/decompression image data using the Wavelet Transformation and Huffman Coding. This technique will be possible to transmit of the geographic information effectively such as position data, attitude data, and image data acquired by Mobile Mapping Systems in the wireless internet environment when data is transmitted in real-time.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.435-436
• /
• 2015

In order to analyze and maximize efficiency of advertise, business put more importance on SNS. Especially, keyword extraction analyses based on Hadoop receive attention. The existing keyword extraction analyses have mostly MapReduce processes. Due to that, it causes problems data base would not update in real time like SNS system. In this study, we indicate limitations of the existing model and suggest new model using Storm technique to analyze data in real time.

• Proceedings of the KIPE Conference
• /
• 2000.07a
• /
• pp.249-252
• /
• 2000

원자력 교육원 #2(KNPEC-2) 시뮬레이터는 1980년도 중반에 웨스팅하우스에 의해 공급되어 계속 사용되어 오다가 현재 성능개선 연구가 진행 중이다. 이번 성능개선을 통해 기존의 컴퓨터 시스템(Gould MPX)와 소프트웨어의 전면 교체가 이루어지고 있으며 최적 계산 코드를 이용한 실시간 열수력 모델 (ARTS; Advanced Real-Time Thermal-Hydraulics Simulation) 개발 , 2-Group 3D 실시간 노심모델(REMARK ; REal Time Multigroup Advanced Reactor Kinetics)를 이용한 노심 주기개선 (Cycle Update) 가상현실 기술 등을 이용한 컴퓨터 교육지원 시스템(CATS: Computer Assister Training System)등 새로운 시도가 이루어지고 있으며 본 논문은 이러한 새로운 시도가 이루어지고 있으며 본 논문은 이러한 새로운 시도들 및 그 결과에 대해 기술하고 있다. 기준발전소(Reference Plant)인 영광 1호기 12주기의 노심모델로 주기개선(Cycle Update)을 위한 REMARK의 입력자료 생성을 위해 핵설계 전산체계인 APA(ALPHA-PHOENIX-ANC) 시스템의 출력으로부터 자동으로 REMARK 입력데이타를 생성하기 위한 GUI툴 개발하였다. 또 이를 이용하여 개발된 노심모델은 최적계산코드(RETRAn 3D) 의 열수력 해법을 이용하여 개발된 NSSS 열수력코드(ARTS) 와 결합(Integration) 되어 안정 및 과도 상태 시험에 사용되었으며 원자로 냉각재 펌프 정지등의 몇 가지 과도 시험 계산결과 기존 해석 결과와 잘 일치하였다 중앙제어실(MCR; Main Control Room)내의 운전원 행동만 훈련하도록 되어있는 기존시뮬레이터의 한계를 극복하기 위해 가상현실 (VR) 저작도구를 이용한 발전소 현장 내부를 표현하는 가상발전소 (Virtual Plant) 발전소 현장에 소재하여 기존 시뮬레이터의 모의한계 밖에 있던 패널을 표현한 가상판넬(Virtual Panel)등과 강의실에서 발전소 모의 훈련을 가능케 하기 위해 가상현실 기술을 이용한 컴퓨터 지원 교육훈력 시스템(CATS ; Computer Assister Training System)을 개발 중이며 일부 개발부분을 소개하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.387-391
• /
• 2008

For the detection of moving objects, background subtraction methods are widely used. In case the background has variation, we need to update the background in real-time for the reliable detection of foreground objects. Gaussian mixture model (GMM) combined with probabilistic learning is one of the most popular methods for the real-time update of the background. However, it does not work well in the complex and dynamic backgrounds with high traffic regions. In this paper, we propose a new method for modelling and updating more reliably the complex and dynamic backgrounds based on the probabilistic learning and the layered processing.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.4
• /
• pp.909-916
• /
• 2010

In ubiquitous GIS services is possible to use the spatio-temporal data using a mobile device at anytime. Also, client is transmitted latest spatio-temporal data from server. But traditional systems have a problem that the time of transmitting latest information from server to client takes long time because of collecting data periodically. In this paper, we proposed Two-way Synchronization system supporting field update to solve the existing problem. This system uses mobile device for collecting changed data in the real world and sending collected data to server.

• Journal of Korean Society of Transportation
• /
• v.24 no.4 s.90
• /
• pp.117-127
• /
• 2006

Many large cities in Korea have implemented or planed to implement a bus information system(BIS) to improve service quality for bus Passengers, mainly by Providing bus arrival time at bus stations. In those systems, similar systematic errors to estimate the bus arrival time occur, which are caused by the cycle time to identify each bus location, the information processing time of the center system, and the cycle time to update the bus arrival information on each terminal. This paper investigated each cause sequentially and estimated three expectations related to the above three causes, respectively using the random incidence concept. Through a validation using real data from a BIS in a city in Korea, fairly amount of improvements on the bus arrival time estimation have been observed.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.8
• /
• pp.855-861
• /
• 2007

Haptic rendering is a process providing force feedback during interactions between a user and a virtual object. This paper presents a real-time haptic rendering technique for deformable objects based on visual information of intervention between a tool and a real object in a remote place. A user can feel the artificial reaction force through a haptic device in real-time when a slave system exerts manipulation tasks on a deformable object. The models of the deformable object and the manipulator are created from the captured image obtained with a CCD camera and the recognition of objects is achieved using image processing techniques. The force at a rate of 1 kHz for stable haptic interaction is deduced using extrapolation of forces at a low update rate. The rendering algorithm developed was tested and validated on a test platform consisting of a one-dimensional indentation device and an off-the shelf force feedback device. This software system can be used in a cellular manipulation system providing artificial force feedback to enhance a success rate of operations.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.3
• /
• pp.121-130
• /
• 2017

NAND flash memory has advantages of non-volatility, little power consumption and fast access time. However, it suffers from inability that does not provide to update-in-place and the erase cycle is limited. Moreover, the unit of read/write operation is a page and the unit of erase operation is a block. Therefore, erase operation is slower than other operations. The AGC, the proposed garbage collection policy focuses on not only garbage collection time reduction for real-time guarantee but also wear-leveling for a flash memory lifetime. In order to achieve above goals, we define three garbage collection operating modes: Fast Mode, Smart Mode, and Wear-leveling Mode. The proposed policy decides the garbage collection mode depending on system CPU usage rate. Fast Mode selects the dirtiest block as victim block to minimize the erase operation time. However, Smart Mode selects the victim block by reflecting the invalid page number and block erase count to minimizing the erase operation time and deviation of block erase count. Wear-leveling Mode operates similar to Smart Mode and it makes groups and relocates the pages which has the similar update time. We implemented the proposed policy and measured the performance compare with the existing policies. Simulation results show that the proposed policy performs better than Cost-benefit policy with the 55% reduction in the operation time. Also, it performs better than Greedy policy with the 87% reduction in the deviation of erase count. Most of all, the proposed policy works adaptively according to the CPU usage rate, and guarantees the real-time performance of the system.