• Journal of the Korean Society of Industry Convergence
• /
• v.22 no.2
• /
• pp.137-146
• /
• 2019

This study proposed a new technique to control of mobile robot based on voice command for (Human-Robot Cooperative operation in manufacturing precess). High performance voice recognition and control system was designed In this paper for smart factory. robust voice recognition is essential for a robot to communicate with people. One of the main problems with voice recognition robots is that robots inevitably effects real environment including with noises. The noise is captured with strong power by the microphones, because the noise sources are closed to the microphones. The signal-to-noise ratio of input voice becomes quite low. However, it is possible to estimate the noise by using information on the robot's own motions and postures, because a type of motion/gesture produces almost the same pattern of noise every time it is performed. In this paper, we describe an robust voice recognition system which can robustly recognize voice by adults and students in noisy environments. It is illustrated by experiments the voice recognition performance of mobile robot placed in a real noisy environment.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1129-1135
• /
• 2020

The effects of battery aging limit the rechargeable capacity, State of Health(SoH). It is very important to estimate the SoH in the battery monitoring system(BMS) and many algorithms of measuring the internal resistance of the battery were proposed. A method is used by applying a current source of a specific frequency to the battery and measuring the voltage response. When charging harmonic noise is generated in the voltage response, it results in poor resistance measurement accuracy. In this paper, a robust battery internal resistance measurement algorithm is proposed to eliminate the effect of charging noise by integrating the current source and voltage response signals for a certain period. It showed excellent accuracy and stable measurement results. Applying to the BMS for uninterruptible power supply, the usefulness of the proposed method is verified.

• Journal of the Chungcheong Mathematical Society
• /
• v.34 no.1
• /
• pp.37-53
• /
• 2021

In this paper we define scaled visual curvature and visual Frenet frame that can be visually accepted for discrete space curves. Scaled visual curvature is relatively simple compared to multi-scale visual curvature and easy to control the influence of noise. We adopt scaled minimizing directions of height functions on each neighborhood. Minimizing direction at a point of a curve is a direction that makes the point a local minimum. Minimizing direction can be given by a small noise around the point. To reduce this kind of influence of noise we exmine the direction whether it makes the point minimum in a neighborhood of some size. If this happens we call the direction scaled minimizing direction of C at p ∈ C in a neighborhood Br(p). Normal vector of a space curve is a second derivative of the curve but we characterize the normal vector of a curve by an integration of minimizing directions. Since integration is more robust to noise, we can find more robust definition of discrete normal vector, visual normal vector. On the other hand, the set of minimizing directions span the normal plane in the case of smooth curve. So we can find the tangent vector from minimizing directions. This lead to the definition of visual tangent vector which is orthogonal to the visual normal vector. By the cross product of visual tangent vector and visual normal vector, we can define visual binormal vector and form a Frenet frame. We examine these concepts to some discrete curve with noise and can see that the scaled visual curvature and visual Frenet frame approximate the original geometric invariants.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.22-34
• /
• 2024

Exposure to varied noisy environments impairs the recognition performance of artificial intelligence-based speech recognition technologies. Degraded-performance services can be utilized as limited systems that assure good performance in certain environments, but impair the general quality of speech recognition services. This study introduces an audiovisual speech recognition (AVSR) model robust to various noise settings, mimicking human dialogue recognition elements. The model converts word embeddings and log-Mel spectrograms into feature vectors for audio recognition. A dense spatial-temporal convolutional neural network model extracts features from log-Mel spectrograms, transformed for visual-based recognition. This approach exhibits improved aural and visual recognition capabilities. We assess the signal-to-noise ratio in nine synthesized noise environments, with the proposed model exhibiting lower average error rates. The error rate for the AVSR model using a three-feature multi-fusion method is 1.711%, compared to the general 3.939% rate. This model is applicable in noise-affected environments owing to its enhanced stability and recognition rate.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.11
• /
• pp.1305-1310
• /
• 2012

The performance of a system can be affected by the variance of noise factors, which arise owing to uncertainties of the material properties and environmental factors acting on the system. For robust design optimization of the system performance, it is necessary to minimize the effect of the variance of the noise factors that are impossible to control. However, present robust design techniques consider the variation of design factors, and not the noise factors, as being important. Furthermore, it is necessary to assume a normal distribution; however, a normal distribution is often not suitable to estimate the variations. In this study, a robust design technique is proposed to consider the variation of noise factors that are estimated as non-normal distributions in a real experiment. As an example of an engineering problem, a deep-sea manganese nodule miner tracked vehicle is used to demonstrate the feasibility of the proposed method.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.180-186
• /
• 2004

The presence of noise in speech signals degrades the performance of recognition systems in which there are mismatches between the training and test environments. To make a speech recognizer robust, it is necessary to compensate these mismatches. In this paper, we studied about an improvement of stochastic feature extraction based on band-SNR for robust speech recognition. At first, we proposed a modified version of the multi-band spectral subtraction (MSS) method which adjusts the subtraction level of noise spectrum according to band-SNR. In the proposed method referred as M-MSS, a noise normalization factor was newly introduced to finely control the over-estimation factor depending on the band-SNR. Also, we modified the architecture of the stochastic feature extraction (SFE) method. We could get a better performance when the spectral subtraction was applied in the power spectrum domain than in the mel-scale domain. This method is denoted as M-SFE. Last, we applied the M-MSS method to the modified stochastic feature extraction structure, which is denoted as the MMSS-MSFE method. The proposed methods were evaluated on isolated word recognition under various noise environments. The average error rates of the M-MSS, M-SFE, and MMSS-MSFE methods over the ordinary spectral subtraction (SS) method were reduced by 18.6%, 15.1%, and 33.9%, respectively. From these results, we can conclude that the proposed methods provide good candidates for robust feature extraction in the noisy speech recognition.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.5
• /
• pp.415-419
• /
• 2008

Detecting and correcting defects on LCD glasses early in the manufacturing process becomes important for panel makers to reduce the manufacturing costs and to improve productivity. Many attempts have been made and were successfully applied to detect and identify simple defects such as scratches, dents, and foreign objects on glasses. However, it is still difficult to robustly detect low-contrast defect region, called Mura or blemish area on glasses. Typically, these defect areas are roughly defined as relatively large, several millimeters of diameter, and relatively dark and/or bright region of low Signal-to-Noise Ratio (SNR) against background of low-frequency signal. The aim of this article is to present a robust algorithm to segment these blemish defects. Early 90's, a highly robust estimator, known as the Model-Fitting (MF) estimator was developed by X. Zhuang et. al. and have been successfully used in many computer vision application. Compared to the conventional Least-Square (LS) estimator the MF estimator can successfully estimate model parameters from a dataset of contaminated Gaussian mixture. Such a noise model is defined as a regular white Gaussian noise model with probability $1-\varepsilon$ plus an outlier process with probability $varepsilon$. In the sense of robust estimation, the blemish defect in images can be considered as being a group of outliers in the process of estimating image background model parameters. The algorithm developed in this paper uses a modified MF estimator to robustly estimate the background model and as a by-product to segment the blemish defects, the outliers.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.10a
• /
• pp.393-394
• /
• 2010

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.04a
• /
• pp.518-519
• /
• 2014

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.483-487
• /
• 1986

The LQG/LTR method is applied to the real servomechanism with the unknown modeling error and system noise variance Q$_{2}$. The equivalent discretized LQG controller is implemented on the 16-bit microcomputer and the experimental results show the improved stability and the satisfactory performance when the noise variance Q$_{2}$ is increased infinitly.