• 대한의용생체공학회:의공학회지
• /
• 제40권1호
• /
• pp.32-37
• /
• 2019

In this study, we designed a blending algorithm for rate adaptive pacing for cardiac pacemaker. Generally, rate adaptive pacing (RAP) is applied to patients whose heart rate does not rise during exercise for chronotropic incompetence (CI) patient. It is very important to develop an algorithm for RAP that can be properly applied to CI patients. In order to design an RAP algorithm we used dual sensors. Firstly, we designed a bio-signal measurement system based on the dual sensors, which are accelerometer and respiratory system. Secondly, we conducted treadmill test for the simulation experiment while using 3-lead ECG as reference. Finally, we designed a blending algorithm based on activation state of the dual sensors. The proposed blending algorithm was subdivided into three sections based on the accelerometer signal, which are rapidly increased section (W1), hardly changed section (W2), and decreased section (W3). Each weight is set aside for each section. To evaluate this algorithm, ten healthy adult males were participated. The correlation and Root Mean Square Error between the proposed algorithm and the reference were compared, and shown to be r=0.88 and 2.82 bpm, respectively. These results show that the proposed blending algorithm of dual sensors enables proper tracking of the heart rate during exercise. Also, it shows the possibility that the proposed blending algorithm can be applied to improve quality of life of the chronotropic incompetence patient.

• 한국기계가공학회지
• /
• 제21권8호
• /
• pp.9-18
• /
• 2022

As the COVID-19 continues, the demand for robotic technology that can be applied in face-to-face tasks such as delivery and transportation, is increasing. Although these technologies have been developed and applied in various industries, the robots can only be operated in a tidy indoor environment and have limitations in terms of payload. To overcome these problems, we developed a 2 degree of freedom(DOF) environmental adaptive robot leg with a double 1-DOF counterbalance mechanism (CBM) based on wire roller. The double 1-DOF CBM is applied to the two revolute joints of the proposed robot leg to compensate for the weight of the mobile robot platform and part of the payload. In addition, the link of the robot leg is designed in a parallelogram structure based on a belt pulley to enable efficient control of the mobile platform. In this study, we propose the principle and structure of the CBM that is suitable for the robot leg, and design of the counterbalance robot leg module for the environment-adaptive control. Further, we verify the performance of the proposed counterbalance robot leg by using dynamic simulations and experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권11호
• /
• pp.4084-4104
• /
• 2021

Visual place recognition (VPR) is a fundamental yet challenging task of mobile robot navigation and localization. The existing VPR methods are usually based on some pairwise similarity of image descriptors, so they are sensitive to visual appearance change and also computationally expensive. This paper proposes a simple yet effective four-step method that achieves adaptive convolutional neural network (CNN) landmark matching for VPR. First, based on the features extracted from existing CNN models, the regions with higher significance scores are selected as landmarks. Then, according to the coordinate positions of potential landmarks, landmark matching is improved by removing mismatched landmark pairs. Finally, considering the significance scores obtained in the first step, robust image retrieval is performed based on adaptive landmark matching, and it gives more weight to the landmark matching pairs with higher significance scores. To verify the efficiency and robustness of the proposed method, evaluations are conducted on standard benchmark datasets. The experimental results indicate that the proposed method reduces the feature representation space of place images by more than 75% with negligible loss in recognition precision. Also, it achieves a fast matching speed in similarity calculation, satisfying the real-time requirement.

• 한국통신학회논문지
• /
• 제28권3A호
• /
• pp.111-118
• /
• 2003

적응형 송신 빔 성형 시스템에서 순방향 링크의 채널 특성을 기지국에 정확히 전달하는 것은 시스템 성능을 결정하는 중요한 요소이다. FDD 방식 시스템의 경우 순방향 채널의 정보는 일반적으로 귀환 채널을 통해 전달되며 송신 안테나 수에 비례하여 증가하게 된다. 이 논문에서는 N개의 송신안테나를 갖는 빔 성형 시스템이 2N개의 송신 안테나 시스템으로 확장되는 반면, 귀환 채널은 귀환 전송 비트율의 제한으로 인해 기존의 N-안테나 시스템의 귀환 채널이 그대로 유지된다. 제한된 귀환 채널 정보의 사용 효율을 높여 시스템의 성능을 향상시키기 위해 적응형 송신 빔 성형 방식과 안테나 선택 다이버시티 방식을 결합한 순방향 링크 CDMA 시스템을 제안하고 모의 실험을 통해 성능을 연구한다. 제한된 귀환 채널 비트를 갖는 시스템에서, 송신 안테나 수와 안테나 선택 방식에 따른 시스템 성능을 주파수 비선택적 페이딩 채널 및 다중 경로 페이딩 채널에서 모의 실험을 통해 정량화 한다. 모의 실험 결과는 송신 안테나 선택 방식을 적용함으로써 각 안테나당 할당되는 정보 비트 수를 증가시켜 양자화로 인한 오류를 줄이고, 선택 다이버시티 이득을 얻음으로써 전체 시스템 성능이 개선됨을 보인다.

• 정보학연구
• /
• 제8권1호
• /
• pp.119-124
• /
• 2005

Learning procedure in the neural network is updating of weights between neurons. Unadequate initial learning coefficient causes excessive iterations of learning process or incorrect learning results and degrades learning efficiency. In this paper, adaptive learning algorithm is proposed to increase the efficient in the learning algorithms of Kohonens Self-Organization Neural networks. The algorithm updates the weights adaptively when learning procedure runs. To prove the efficiency the algorithm is experimented to clustering of the random weight. The result shows improved learning rate about 42~55% ; less iteration counts with correct answer.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.189-192
• /
• 2001

In this paper, we propose a ADALINE controller using fuzzy-backpropagation algorithm to adjust weight. In the proposed ADALINE controller, using fuzzy algorithm for traning neural network, controller make use of ADALINE due to simple and computing efficiency. This controller includes adaptive learning rate to accelerate teaming. It applies to servo-motor as an controlled process. And then it take a simulation for the position control, so the verify the usefulness of the proposed ADALINE controller.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 제14회 신호처리 합동 학술대회 논문집
• /
• pp.915-918
• /
• 2001

In this paper we study a new convergence behavior of the least mean fourth (LMF) algorithm where the error raised to the power of four is minimized for a multiple sinusoidal input and Gaussian measurement noise. Here we newly obtain the convergence equation for the sum of the mean of the squared weight errors, which indicates that the transient behavior can differ depending on the relative sizes of the Gaussian noise and the convergence constant. It should be noted that no similar results can be expected from the previous analysis by Walach and Widrow.

• 한국통신학회논문지
• /
• 제26권12A호
• /
• pp.2043-2049
• /
• 2001

In this paper we study the convergence behavior of the least mean fourth(LMF) algorithm where the error raised to the power of four is minimized for a multiple sinusoidal input and Gaussian measurement noise. Here we newly obtain the convergence equation for the sum of the mean of the squared weight errors, which indicates that the transient behavior can differ depending on the relative sizes of the Gaussian noise and the convergence constant. It should be noted that no similar results can be expected from the previous analysis by Walach add Widrow.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.841-842
• /
• 2008

스테레오 매칭(Stereo Matching) 기법에 대한 전역적인 방법과 지역적인 방법에 대한 연구가 활발하게 진행되고 있다. 최근의 적응적 영역 가중치 방법(Adaptive Support-Weight)은 매우 뛰어난 결과에 비해 많은 계산 시간이 필요하다. 따라서 로봇시스템에서 스테레오 매칭을 이용하기에는 부적합하다. 본 논문에서는 분리 가능한 Bilateral 필터를 이용하여 빠른 스테레오매칭 기법을 제안한다

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.484-489
• /
• 1992

The inference of fuzzy controller can be considered a mapping from the controller input to membership value. The membership value, a kind of weight, has a role to decide if the input is appropriate to the rule. The membership function is described by several values, which are decided by a learning method. The learning method is adopted from adaptive filtering theory. The simulation shows the proposed fuzzy controller can learn linear and nonlinear functions. the structure of the proposed fuzzy controller becomes a kind of neural network.