• 제어로봇시스템학회논문지
• /
• 제10권8호
• /
• pp.696-702
• /
• 2004

An optimal trajectory generation algorithm for capturing a moving object by a mobile robot in real-time is proposed in this paper. The linear and rotational velocities of the moving object are estimated using the Kalman filter, as a state estimator. For the estimation, the moving object is tracked by a 2-DOF active camera mounted on the mobile robot, which enables a mobile manipulator to track the mobile robot until the capturing moment. The optimal trajectory for capturing the moving object is dependent on the initial conditions of the mobile robot as well as the moving object. Therefore, real-time trajectory planning for the mobile robot is definitely required for the successful capturing of the moving object. The performance of proposed algorithm is verified through the real experiments and the superiority is demonstrated by comparing to other algorithms.

• Journal of Electrical Engineering and Technology
• /
• 제13권4호
• /
• pp.1419-1424
• /
• 2018

Very short-term electric load forecasting is essential for real-time power system operation. In this paper, a very short-term electric load forecasting technique applying the Kalman filter algorithm is proposed. In order to apply the Kalman filter algorithm to electric load forecasting, an electrical load forecasting algorithm is defined as an observation model and a state space model in a time domain. In addition, in order to precisely reflect the noise characteristics of the Kalman filter algorithm, the optimal error covariance matrixes Q and R are selected from several experiments. The proposed algorithm is expected to contribute to stable real-time power system operation by providing a precise electric load forecasting result in the next six hours.

• 대한음성학회지:말소리
• /
• 제54호
• /
• pp.63-72
• /
• 2005

The performance of speech recognition system is degraded by the mismatch between training and test environments. Many studies have been presented to compensate for noise components in the cepstral domain. Recently, higher order cepstral moment normalization method has been introduced to improve recognition accuracy. In this paper, we present real-time high order moment normalization method with post-processing smoothing filter to reduce the parameter estimation error in higher order moment computation. In experiments using Aurora2 database, we obtained error rate reduction of 44.7% with proposed algorithm in comparison with baseline system.

• 대한임베디드공학회논문지
• /
• 제16권6호
• /
• pp.247-251
• /
• 2021

We propose a protocol for Ethernet-based industrial real-time communication networks. In the protocol, the master periodically transmits control frames to all slaves, and the ring-type network topology is selected to achieve high-speed transmission speed. The proposed protocol is implemented in the form of both firmware and Linux kernel modules. To improve the transmission speed, the MAC address table is disabled in the firmware implementation, and the NAPI function of the Ethernet driver is removed in the Linux kernel module implementation. A network experiment environment is built with four ARM processor-based embedded systems and network operation experiments are performed for various frame sizes. From the experimental results, it is verified that the proposed protocol normally operates, and the firmware implementation shows better transmission speed than the Linux kernel module implementation.

• 한국멀티미디어학회논문지
• /
• 제22권2호
• /
• pp.203-209
• /
• 2019

The performance of an intelligent classifier for detecting malwares added to multimedia contents based on machine learning is highly dependent on the properties of feature set. Especially, in order to determine the malicious code in real time the size of feature set should be as short as possible without reducing the accuracy. In this paper, we introduce an optimal feature selection method to satisfy both high detection rate and the minimum length of feature set against the feature set provided by PEFeatureExtractor well known as a feature extraction tool. For the evaluation of the proposed method, we perform the experiments using Windows Portable Executables 32bits.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2148-2161
• /
• 2019

In this paper, we present a real-time cattle action recognition algorithm to detect the estrus phase of cattle from a live video stream. In order to classify cattle movement, specifically, to detect the mounting action, the most observable sign of the estrus phase, a simple yet effective feature description exploiting motion history images (MHI) is designed. By learning the proposed features using the support vector machine framework, various representative cattle actions, such as mounting, walking, tail wagging, and foot stamping, can be recognized robustly in complex scenes. Thanks to low complexity of the proposed action recognition algorithm, multiple cattle in three enclosures can be monitored simultaneously using a single fisheye camera. Through extensive experiments with real video streams, we confirmed that the proposed algorithm outperforms a conventional human action recognition algorithm by 18% in terms of recognition accuracy even with much smaller dimensional feature description.

• 한국인터넷방송통신학회논문지
• /
• 제9권5호
• /
• pp.85-91
• /
• 2009

IT기술의 발전과 함께 유비쿼터스 (Ubiquitous)환경이 도래하여, 무선 센서 네트워크(Wireless Sensor Networks)기반 연구가 다방면에서 활발히 진행 중에 있다. 본 논문에서는 무선 센서 네트워크에서 센서간 거리를 전방위로 측정할 수 있는 간단한 방안에 대해 제안한다. 그리고 이를 직접 설계하고 구현하여 실험을 통해 검증해 본다. 전방향성 장치를 이용하면 거리 측정 최대 거리는 줄어드는 단점이 있지만 $360^{\circ}$ 방향을 측정 할 수 있다는 장점이 있다. 본 논문에서는 실제 환경에서 실험을 통하여 전방향성 장치가 존재할 경우와 그렇지 않은 경우에 대하여 초음파를 통한 거리측정 방식의 누락율 및 정확도를 분석해 본다.

• Journal of Electrical Engineering and Technology
• /
• 제8권6호
• /
• pp.1431-1438
• /
• 2013

A pack-level battery hardware-in-the-loop simulation (B-HILS) platform is implemented. It consists of dynamic vehicle models using PSAT and multiple control interfaces including real-time 3D driving and GPS mode. In real-time 3D driving mode, user can drive a virtual vehicle using actual drive equipment such as steering wheel and accelerator to generate the cycle profile of the battery. In GPS mode, actual road traffic and terrain effects can be simulated using GPS data while the trajectory is displayed on Google map. In the latter part of the paper, several performance tests of an actual lithium-polymer battery pack are carried out utilizing the developed system. All experiments are conducted as parts of actual development process of a commercial battery pack adopting 2nd generation Prius as a target vehicle model. Through the experiments, the low temperature performance and fuel efficiency of the battery are quantitatively investigated in comparison with the original nickel-metal hydride (NiMH) pack of the Prius.

• 로봇학회논문지
• /
• 제2권1호
• /
• pp.1-8
• /
• 2007

This paper presents a PID automatic gain-tuning algorithm for the electronic throttle valve which is driven by wire. Since the system characteristics of position control for electronic throttle valve are so complicated that both the real time robustness and the manufacturing cost must be considered for mass production. To resolve this paradox, a kind of algorithm called RLS (Recursive Least Square) is adopted for the control of the ETB (Electronic Throttle Body). Using this algorithm, the PID gains can be adjusted automatically with the estimated system parameters. Furthermore, a pre-filter is supplemented for the sake of the robustness against the friction and loads. From the industrial requests for the system, the design specifications are decided as follows: the settling time should be less than 1sec and the overshoot should be kept below 3%. The results of the experiments based on this approach show that the high robustness can be achieved while the system stability is satisfied steadily. A parameter estimation scheme and a gain-tuning algorithm have been properly combined and utilized in this research and the effectiveness is verified through the real experiments.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1972-1977
• /
• 2005

In this work, an inverse dynamic control method is developed to enhance tracking performance of industrial robots, which effectively deal with the nonlinear dynamic interferential forces. In general, the DFF (Dynamic Feed-Forward) controller and the CTM (Computed-Torque Method) controller are used for dynamic control for industrial robots. We study on the practical issues for implementing these inverse dynamic controllers via simulations and experiments. We develop the dynamic models in two different ways. One is a model designed through Newton-Euler method for real time computation and the other is a model designed through SimMechanics for evaluating the developed controller via simulations. We evaluate the nominal performance and robustness of the controller via simulations and experiments using serial 4-DOF HyRoHILS (Hyundai Robot Hardware-In-the-Loop Simulation) system. The results show that the inverse dynamic controller is effective and practically useful for a real control structure.