• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.748-757
• /
• 2021

In this paper, we propose a gaze estimation network in which eye landmark position detection and gaze direction vector estimation are integrated into one deep learning network. The proposed network uses the Stacked Hourglass Network as a backbone structure and is largely composed of three parts: a landmark detector, a feature map extractor, and a gaze direction estimator. The landmark detector estimates the coordinates of 50 eye landmarks, and the feature map extractor generates a feature map of the eye image for estimating the gaze direction. And the gaze direction estimator estimates the final gaze direction vector by combining each output result. The proposed network was trained using virtual synthetic eye images and landmark coordinate data generated through the UnityEyes dataset, and the MPIIGaze dataset consisting of real human eye images was used for performance evaluation. Through the experiment, the gaze estimation error showed a performance of 3.9, and the estimation speed of the network was 42 FPS (Frames per second).

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.10
• /
• pp.3873-3879
• /
• 2010

Home service robot are not working in the fixed task such as industrial robot, because they are together with human in the same indoor space, but have to do in much more flexible and various environments. Most of them are developed on the base of the wheel-base mobile robot in the same method as a vehicle robot for factory automation. In these days, for holonomic system characteristics, omni-directional wheels are used in the mobile robot. A holonomicrobot, using omni-directional wheels, is capable of driving in any direction. But trajectory control for omni-directional mobile robot is not easy. Especially, azimuth control which sensor uncertainty problem is included is much more difficult. This paper develops trajectory controller of 3-wheels omni-directional mobile robot using fuzzy azimuth estimator. A trajectory controller for an omni-directional mobile robot, which each motor is controlled by an individual PID law to follow the speed command from inverse kinematics, needs a precise sensing data of its azimuth and exact estimation of reference azimuth value. It has imprecision and uncertainty inherent to perception sensors for azimuth. In this paper, they are solved by using fuzzy logic inference which can be used straightforward to perform the control of the mobile robot by means of the fuzzy behavior-based scheme already existent in literature. Finally, the good performance of the developed mobile robot is confirmed through live tests of path control task.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.6
• /
• pp.855-864
• /
• 2011

In this paper, a nonlinear controller based on adaptive back-stepping method is proposed for high performance operation of IPMSM(Interior Permanent Magnet Synchronous Motor). First, in order to improve the performance of speed tracking a nonlinear back-stepping controller is designed. Since it is difficult to control the high performance driving without considering parameter variation, a parameter estimator is included to adapt to the variation of load torque in real time. In addition, for the efficiency of power consumption of the motor, controller is designed to operate motor with minimum current for maximum torque. The proposed controller is applied through simulation to the a 2-hp IPMSM for the angular velocity reference tracking performance and load torque volatility estimation, and to test the MTPA(Maximum Torque per Ampere) operation in constant torque operation region. The result verifies the efficacy of the proposed controller.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.8C
• /
• pp.806-813
• /
• 2009

This paper presents a modem chip design for high-speed wireless communications. Among the high-speed communication technologies, we design the UWB (Ultra-Wideband) modem SoC (System-on-Chip) Chip based on a MB-OFDM scheme which uses wide frequency band and gives low frequency interference to other communication services. The baseband system of the modem SoC chip is designed according to the standard document published by WiMedia. The SoC chip consists of FFT/IFFT (Fast Fourier Transform/Inverse Fast Fourier Transform), transmitter, receiver, symbol synchronizer, frequency offset estimator, Viterbi decoder, and other receiving parts. The chip is designed using 90nm CMOS (Complementary Metal-Oxide-Semiconductor) procedure. The chip size is about 5mm x 5mm and was fab-out in July 20th, 2009.

• Journal of The Institute of Information and Telecommunication Facilities Engineering
• /
• v.1 no.2
• /
• pp.4-18
• /
• 2002

In this paper, we propose the new design method for the OFDM based modem that is considerd a standard of wireless communication in indoor environments. We designed a improved FFT/IFFT in order to satisfy a data rate $6{\sim}54$Mbps required homenetworking of high speed and a improved channel equalization circuit using pilot signals for modile environments. And we designed a carrier offset estimator that uses the $tan^{-1}$ circuit to organize a memory structure. All steps are verifed performance through a FPGA and are implemented ASIC to use a standard library cell.

• The Korean Journal of Applied Statistics
• /
• v.31 no.4
• /
• pp.463-473
• /
• 2018

Variable selection algorithms for linear regression models of large data are considered. Many algorithms are proposed focusing on the speed and the robustness of algorithms. Among them variance inflation factor (VIF) regression is fast and accurate due to the use of a streamwise regression approach. But a VIF regression is susceptible to outliers because it estimates a model by a least-square method. A robust criterion using a weighted estimator has been proposed for the robustness of algorithm; in addition, a robust VIF regression has also been proposed for the same purpose. In this article a fast and robust variable selection method is suggested via a VIF regression with detecting and removing potential outliers. A simulation study and an analysis of a dataset are conducted to compare the suggested method with other methods.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.10
• /
• pp.1308-1316
• /
• 2018

Accurate tuning of parameter is very important in vector-controlled induction motor. Among the parameters of induction motor, detuning of rotor resistance used in controller design deteriorates drive performance. This paper presents a novel rotor resistance estimation strategy using slip angular velocity in vector-controlled induction motor drives. The slip angular velocity can be calculated by two methods. Firstly, it can be induced from the rotor voltage equation. Secondly, it can be induced from the difference between synchronous angular velocity and rotor angular velocity. The first method includes the rotor resistance, while the second method dose not include this parameter. From this fact, the rotor resistance can be identified by comparing the slip angular velocities in the two methods. In the tuned states of the rotor resistance, performances of flux estimator and speed drive are discussed. The simulation and experimental results are given to verify the validity of the proposed method in various situations.

• Proceedings of the KIPE Conference
• /
• 2001.12a
• /
• pp.23-26
• /
• 2001

This paper presents a digital motion control system for Reluctance Synchronous Motor (RSM) drives with direct torque control (DTC). The system consists of stator flux observer, torque estimator: two hysteresis band controllers, an optimal switching look-up table, IGBT voltage source inverter(VSI), and TMS320C31 DSP controller by using fully integrated control software. The stator flux observer is based on the combined voltage and current model with stator flux feedback adaptive control of which inputs are current, voltage and actual rotor angle for wide speed range. In order to prove the suggested motion control algorithm, There are some simulation and testing at actual experimental system. The developed digitally high-performance motion control system are shown a good motion control response characteristic results and high performance features using 1.0Kw RSM.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.4
• /
• pp.730-738
• /
• 2007

As well known when the linear machine is operated between two points repeatedly under positioning control, there are various positioning error at the moment of zero speed owing to the non-linear disturbance like as unpredictable friction force. To remove this positioning error, a simple least order disturbance observer is introduced and is actually implemented in this study. Due to this simple algorithm the over-all machine system can be modified to simple arbitrary given one-mass load without any disturbance. So, the total construction process for positioning control system is much easier than old one. Moreover, to generate a proper effective position profile with the limited actual machine force, a very powerful on-line mass identification algorithm using the load force estimator is presented. In the proposed mass identification algorithm, the exact load mass can be calculated during only one moving stage under a normally generated position profile. All presented algorithm is verified with experimental result with commercial linear servo machine system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.6A
• /
• pp.660-666
• /
• 2004

Transmitted data may be compensated by using estimated channel values that are obtained with pilot symbols in OFDM-CDMA systems. Generally, a USE (Minimum Mean-Squared Error) estimator using correlations between pilot symbols gives good results, but its structure is so complicated. Starting with a modification of PA (Pilot-Aided) algorithm using pilot symbols and PADD (Pilot-Aided Decision-Directed) algorithm using both pilot and data symbols, a new channel estimation algorithm with more simpler structure is proposed. The performance of this algorithm is evaluated with varying mobile speed in a Ralyleigh multipath fading environment through computer simulations. The simulation results show that the proposed channel estimation algorithm outperforms a conventional PA algorithm.