• Journal of Korea Multimedia Society
• /
• v.24 no.5
• /
• pp.676-683
• /
• 2021

Wi-Fi Round-trip timing (RTT) based location estimation technology estimates the distance between the user and the AP based on the transmission and reception time of the signal. This is because reception instability and signal distortion are greater than that of a Received Signal Strength Indicator (RSSI) based fingerprint in an indoor NLOS environment, resulting in a large position error due to multipath fading. To solve this problem, in this paper, we propose an end-to-end based WiFi Trilateration Net (WTN) that combines neural network-based RTT correction and trilateral positioning network, respectively. The proposed WTN is composed of an RNN-based correction network to improve the RTT distance accuracy and a neural network-based trilateral positioning network for real-time positioning implemented in an end-to-end structure. The proposed network improves learning efficiency by changing the trilateral positioning algorithm, which cannot be learned through differentiation due to mathematical operations, to a neural network. In addition, in order to increase the stability of the TOA based RTT, a correction network is applied in the scanning step to collect reliable distance estimation values from each RTT AP.

• Electronics and Telecommunications Trends
• /
• v.34 no.4
• /
• pp.55-64
• /
• 2019

In this study, we introduce trends in neural-network-based deep learning research applied to dialogue systems. Recently, end-to-end trainable goal-oriented dialogue systems using long short-term memory, sequence-to-sequence models, among others, have been studied to overcome the difficulties of domain adaptation and error recognition and recovery in traditional pipeline goal-oriented dialogue systems. In addition, some research has been conducted on applying reinforcement learning to end-to-end trainable goal-oriented dialogue systems to learn dialogue strategies that do not appear in training corpora. Recent neural network models for end-to-end trainable chit-chat systems have been improved using dialogue context as well as personal and topic information to produce a more natural human conversation. Unlike previous studies that have applied different approaches to goal-oriented dialogue systems and chit-chat systems respectively, recent studies have attempted to apply end-to-end trainable approaches based on deep neural networks in common to them. Acquiring dialogue corpora for training is now necessary. Therefore, future research will focus on easily and cheaply acquiring dialogue corpora and training with small annotated dialogue corpora and/or large raw dialogues.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.12
• /
• pp.143-152
• /
• 1997

The deflection of an end mill is very important in machining process and cutting simulation because it affects directly workpiece accuracy, cutting force, and chattering. In this study, the deflection of the end mill was studied both experimentally and by using finite element analysis. And the moment of inertia of cross sections of the helical end mill is calculated for the determination of the relation between geometry of radial cross section and rigidity of the tools. Using the Bernoulli-Euler beam theory and the concept of equivalent diameter, a deflection model is established, which includes most influences from tool geomety parameters. It was found that helix angle attenuates the rigidity of the end mill by the finite element analysis. As a result, the equivalent diameter is determined by tooth number, inscribed diameter ratio, cross sectional geometry and helix angle. Because the relation betweem equivalent diameter and each factor is nonlinear, neural network is used to decide the equivalent diameter. Input patterns and desired outputs for the neural network are obtained by FEM analysis in several case of end milling operations.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.7
• /
• pp.728-733
• /
• 2009

This paper presents the compensation of mechanical deflection error in SCARA robot. End of robot gripper is deflected by weight of arm and pay-load. If end of robot gripper is deflected constantly regardless of robot configuration, it is not necessary to consider above mechanical deflection error. However, deflection in end of gripper varies because that moment of each axis varies when robot moves, it affects the relative accuracy. I propose the compensation method of deflection error using neural network. FEM analysis to obtain the deflection of gripper end was carried out on various joint angle, the results is used in neural network teaming. The result by simulation showed that maximum relative accuracy reduced maximum 9.48% on a given working area.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.12 no.5
• /
• pp.1-7
• /
• 2003

In order to predict vibrations occurred during end-milling processes, the cutting dynamics was modelled by using neural network and combined with structural dynamics by considering dynamic cutting state. Specific cutting force constants of the cutting dynamics model were obtained by averaging cutting forces. Tool diameter, cutting speed, fled, axial and radial depth of cut were considered as machining factors in neural network model of cutting dynamics. Cutting farces by test and by neural network simulation were compared and the vibration displacement during end-milling was simulated.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.24-31
• /
• 2020

In this paper, the deep convolutional neural network for sound event detection is studied. Especially, the end-to-end neural network, which generates the detection results from the input audio waveform, is studied for weakly supervised problem that includes weakly-labeled and unlabeled dataset. The proposed system is based on the network structure that consists of deeply-stacked 1-dimensional convolutional neural networks, and enhanced by the skip connection and gating mechanism. Additionally, the proposed system is enhanced by the sound event detection and post processings, and the training step using the mean-teacher model is added to deal with the weakly supervised data. The proposed system was evaluated by the Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 Task 4 dataset, and the result shows that the proposed system has F₁-scores of 54 % (segment-based) and 32 % (event-based).

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.272-277
• /
• 2003

In order to predict vibration during end-milling process, the cutting dynamics was modelled by using neural network and combined with structural dynamics by considering dynamic cutting states. Specific cutting constants of the cutting dynamics model were obtained by averaging cutting forces and tool diameter, cutting speed, feed, axial depth radial depth were considered as machining factors. Cutting farces by test and by neural network simulation were compared and the vibration during end-milling was simulated.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.453-458
• /
• 2018

본 논문은 한국어 자소를 인식 단위로 사용한 hybrid CTC-Attention 모델 기반 end-to-end speech recognition을 제안한다. End-to-end speech recognition은 기존에 사용된 DNN-HMM 기반 음향 모델과 N-gram 기반 언어 모델, WFST를 이용한 decoding network라는 여러 개의 모듈로 이루어진 과정을 하나의 DNN network를 통해 처리하는 방법을 말한다. 본 논문에서는 end-to-end 모델의 출력을 추정하기 위해 자소 단위의 출력구조를 사용한다. 자소 기반으로 네트워크를 구성하는 경우, 추정해야 하는 출력 파라미터의 개수가 11,172개에서 49개로 줄어들어 보다 효율적인 학습이 가능하다. 이를 구현하기 위해, end-to-end 학습에 주로 사용되는 DNN 네트워크 구조인 CTC와 Attention network 모델을 조합하여 end-to-end 모델을 구성하였다. 실험 결과, 음절 오류율 기준 10.05%의 성능을 보였다.

• Journal of Mechanical Science and Technology
• /
• v.14 no.6
• /
• pp.622-632
• /
• 2000

In a high precision vertical machining center, the estimation of cutting forces is important for many reasons such as prediction of chatter vibration, surface roughness and so on. The cutting forces are difficult to predict because they are very complex and time variant. In order to predict the cutting forces of end-milling processes for various cutting conditions, their mathematical model is important and the model is based on chip load, cutting geometry, and the relationship between cutting forces and chip loads. Specific cutting force coefficients of the model have been obtained as interpolation function types by averaging forces of cutting tests. In this paper the coefficients are obtained by neural network and the results of the conventional method and those of the proposed method are compared. The results show that the neural network method gives more correct values than the function type and that in the learning stage as the omitted number of experimental data increase the average errors increase as well.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.57-59
• /
• 2017

대화 시스템(dialogue system)은 텍스트나 음성을 통해 다양한 분야에서 특정한 목적을 수행할 수 있는 시스템이다. 대화 시스템을 구현하기 위한 방법으로 인공 신경망(neural network)을 기반으로한 end-to-end learning 방식이 제안되었다. End-to-end learning 방식을 이용한 식당 예약 시스템 모델의 학습을 위해 페이스북은 영어로 이루어진 식당 예약에 관련된 학습 대화 데이터셋(The 6 dialog bAbI tasks)을 구축하였다. 하지만 end-to-end learning 방식의 학습은 많은 학습 데이터가 필요하다는 단점이 존재하는데, 액션 템플릿(action template)의 정의를 통해 도메인 지식을 표현함으로써 일반적인 end-to-end learning 방식보다 적은 학습량으로 좋은 성능의 모델을 학습할 수 있는 Hybrid Code Network 구조를 제안한 연구가 있다. 본 논문에서는 Hybrid Code Network 구조를 이용하여 한국어 식당 예약 시스템을 구축할 수 있는 방법을 제안하고, 한국어로 이루어진 식당 예약에 관련한 학습 대화 데이터를 구축하는 방법을 제안한다.