• Proceedings of the Korea Contents Association Conference
• /
• 2005.11a
• /
• pp.364-370
• /
• 2005

LBL(Layer-By-Layer) algorithms have been proposed to accelerate the training speed of MLPs(Multilayer Perceptrons). In this LBL algorithms, each layer needs a error function for optimization. Especially, error function for hidden layer has a great effect to achieve good performance. In this sense, this paper proposes a new hidden layer error function for improving the performance of LBL algorithm for MLPs. The hidden layer error function is derived from the mean squared error of output layer. Effectiveness of the proposed error function was demonstrated for a handwritten digit recognition and an isolated-word recognition tasks and very fast learning convergence was obtained.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.31-36
• /
• 2014

A study on the indoor positioning system has been active recently for the location based service indoors. In the 3 dimensional positioning system based on the acoustic signal and TDOA technology, the error characteristics of the estimated source position would be changed depending on the number of microphones and the pattern of the microphone array. In this paper, the estimated position error according to the measured distance error between the microphones and the signal source is analyzed, and the optimum microphone array is decided considering the estimated position error patterns and the total amount of the estimated position error.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.54-56
• /
• 2012

본 논문에서는 위치인식 센서기반 무인이송차량(AGV)의 이동 제어 문제를 다루고자 한다. AGV의 진행 경로를 지시하는 랜드마크를 부착할 때 사각지역(dead zone) 및 중첩 지역(overlap zone)이 존재할 경우 위치 추정 오차가 허용 범위를 크게 벗어나게 되며, 이로 인해 AGV가 오동작하는 현상이 발생한다. 이를 해결하기 위해 본 논문에서는 단일 랜드마크 대신 이웃한 2개의 랜드마크 인식을 통해 위치 추정 오차를 보정하는 방안을 제안하였다. 또한, 회전 구간에서 AGV가 방향 전환 직후 지정된 경로에 허용오차 범위 이내로 진입하도록 안쪽 바퀴와 바깥쪽 바퀴의 가속도 제어 알고리즘을 제안하였다. 본 논문에서 개발된 시스템은 화장장 시신 운구용 AGV에 적용하였다. 화장장은 기존 산업 현장에 비해 이동 공간이 협소할 뿐만 아니라 그 특성상 정밀 제어가 필요한 환경이다. 본 논문에서 제안한 방식은 모의 차량에 적용하여 그 타당성을 검증하였으며, 실제 국내 화장장에 AGV 시스템을 적용한 결과 허용오차 범위 이내에서 정상 동작함을 확인하였다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.12
• /
• pp.44-53
• /
• 2007

In this paper, we focus on the developments of complex location awareness algorithms for real-time location based service and precise/stable localization in the outdoor. In the case of using galileo satellite system along with GPS, several error factor such as the ionosphere can be reduced for an increment of used frequency and visible satellites. Therefore, localization estimation error is no longer having problems with location awareness. But, chips synchronization error induces the error of acquisition and tracking, and the performance of receiver can be decreased. In order to solve this problem, this paper proposes a correlator for performance improvement of receiver in the precise localization.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.11
• /
• pp.1442-1447
• /
• 2018

Since the indoor location recognition system using RFID is a method for predicting the indoor position, an error occurs due to the surrounding environment such as an obstacle. In this paper, we plan to reduce errors using back propagation neural networks. The neural network adjusts and trains the connection values between the layers to reduce the error between the actual position of the object with the reader and the expected position of the object through the experiment. In this paper, we propose a method that uses the median method and the radiation method as input to the neural network. Among the two methods, we want to find out which method is more effective in recognizing the actual position in an environment with obstacles and reduce the error. Consequently, the method using the median has less error, and we confirmed that the more the number of data, the smaller the error.

• The Journal of the Korea Contents Association
• /
• v.5 no.6
• /
• pp.57-64
• /
• 2005

LBL(Layer-By-Layer) algorithms have been proposed to accelerate the training speed of MLPs(Multilayer Perceptrons). In this LBL algorithms, each layer needs a error function for optimization. Especially, error function for hidden layer has a great effect to achieve good performance. In this sense, this paper proposes a new hidden layer error function for improving the performance of LBL algorithm for MLPs. The hidden layer error function is derived from the mean squared error of output layer. Effectiveness of the proposed error function was demonstrated for a handwritten digit recognition and an isolated-word recognition tasks and very fast learning convergence was obtained.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1754-1755
• /
• 2007

영구자석을 사용한 마커가 설치된 자기장 이동 경로에서 이동 로봇의 경로 인식을 위한 위치인식 시스템이다. 저가의 홀센서를 일정 간격의 열로 배치한 센서를 설계 제작하고, 그 센서를 이용하여 측정된 마커의 자계 값을 이용하여 경로상의 위치를 인식할 수 있도록 하였다. 마커의 극성을 인식하는 새로운 방법을 제시하고, 인식된 극성을 코드화하여 경로상의 특이점들에 배치함으로써 경로 이동 중에 그 점들의 위치를 인식하도록 하였으며, 로봇은 그 위치에서 미리 정해진 행동을 하도록 프로그램 하였다. 센서들 간의 통신 방법을 논하였고, 각 센서들에서 얻어진 정보로부터 이동 로봇의 경로상의 위치인 측방향 변위 오차와 방향각 오차를 기하학적으로 산출한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.12
• /
• pp.2365-2370
• /
• 2017

In this paper, we propose a motion and position recognition technique considering an experiential VR environment. Motion recognition attaches a plurality of AHRS devices to a body part and defines a coordinate system based on this. Based on the 9 axis motion information measured from each AHRS device, the user's motion is recognized and the motion angle is corrected by extracting the joint angle between the body segments. The location recognition extracts the walking information from the inertial sensor of the AHRS device, recognizes the relative position, and corrects the cumulative error using the BLE fingerprint. To realize the proposed motion and position recognition technique, AHRS-based position recognition and joint angle extraction test were performed. The average error of the position recognition test was 0.25m and the average error of the joint angle extraction test was $3.2^{\circ}$.

• Journal of Korea Multimedia Society
• /
• v.14 no.5
• /
• pp.632-641
• /
• 2011

This paper presents a position error recognition system when the wafer is mounted in cleaning equipment among the wafer manufacturing processes. The proposed system is to enhance the performance in cost and reliability by preventing the wafer cleaning system from damaging by alerting it when it is put in correct position. The proposed algorithm is in obtaining a mapping function from camera and physical wafer by designing and manufacturing the radial shape calibrator to reduce the error by using the conventional chess board one. The system is to install in-line process using high reliable and high accurate position recognition. The experimental results show that the performance of the proposed system is better than that of the existing method for detecting errors within tolerance.

• Proceedings of the KAIS Fall Conference
• /
• 2010.05a
• /
• pp.85-88
• /
• 2010

본 논문은 센서네트워크 환경에 설치된 전력선통신 기반 영상 송수신 장치의 영상정보를 수신하는 기능과, CSS기반 SDS-TWR Ranging을 지원하는 센서모듈로부터 거리정보를 수신하여 제안하는 위치인식 알고리즘을 적용시키고 이동노드의 위치정보를 보여주는 모니터링 기능을 개발하였다. 전력선통신 기반 영상 송수신 장치는 장소의 제약 없이 전기 플러그가 있는 곳이면 사용이 가능한 장점이 있고, 주변의 상황을 확인하기 어려운 센서모듈의 단점을 보완하였다. 그리고 제안한 위치인식 보정 알고리즘은 SDS-TWR의 위치인식 오차 116.8cm를 80.5cm로 위치인식의 성능을 향상시켰으며, 대부분의 위치인식 오차가 1m 내외로 보정함을 확인하였다. 본 논문에서 제시하는 모니터링 기능은 센서 네트워크 및 다양한 환경에서 적용이 가능하여 위치인식 시스템의 서비스 품질의 향상에 도움을 줄 것으로 판단된다.