• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권10호
• /
• pp.5112-5128
• /
• 2016

Tracking human body is an important problem in computer vision field. Tracking failures caused by occlusion can lead to wrong rectification of the target position. In this paper, a robust human tracking algorithm is proposed to address the problem of occlusion, rotation and improve the tracking accuracy. It is based on Tracking-Learning-Detection framework. The key auxiliary information is used in the framework which motivated by the fact that a tracking target is usually embedded in the context that provides useful information. First, face localization method is utilized to find key face location information. Second, the relative position relationship is established between the auxiliary information and the target location. With the relevant model, the key face information will get the current target position when a target has disappeared. Thus, the target can be stably tracked even when it is partially or fully occluded. Experiments are conducted in various challenging videos. In conjunction with online update, the results demonstrate that the proposed method outperforms the traditional TLD algorithm, and it has a relatively better tracking performance than other state-of-the-art methods.

• Journal of Astronomy and Space Sciences
• /
• 제18권1호
• /
• pp.63-70
• /
• 2001

IDGPS(Inverted Differential Global Positioning System)는 GPS의 위치 결정 정밀도를 향상시키는 여러 기법들 중 하나로서 주로 차량 위치 추적이나 물류 시스템에 이용된다. 본 연구에 적용된 IDGPS 기법은 사용자측에서 보내준 GPS 위치해와 위성정보, 그리고 기준국에서 생성된 오차 보정 정보를 이용해 관제국에서 사용자의 위치를 보정하는 것이다. 기준국과 사용자에서 발생한 오차들의 공간적 비상관성(spatial decorrelation)으로 인해 생길 수 있는 정밀도 보정 효과의 저하를 여러 개의 기준국을 이용한 NIDGPS(Network IDGPS)를 통해 해결하였다. 차량항법용 수신기를 사용자 수신기로서 사용하여 실험한 결과, IDGPS와 NIDGPS를 통해 단독 측위에 비해 평균적으로 2배 이상 정밀도가 향상됨을 확인할 수 있었다.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 정기총회 및 추계학술대회 논문집
• /
• pp.3086-3091
• /
• 2011

Many types of door system are used in the Electric Multiple Unit (EMU). The door system is categorized to the two types of door system, sliding door and plug door. The sliding door is widely used in the EMU but has weak point for the noise problem. The plug door has advantages for the noise reduction characteristic. Nowaday electric door system is widely used and motor is a main part of driving system in the door system. We have to know the exact driving position of door system to control proper operation of the door. Actually encoder or tachometer is used for the position controlling, but it makes price increase and failure increase. In this paper, we study and describe motor current calculation method to contol and know exact driving position of door system.

• 제어로봇시스템학회논문지
• /
• 제19권2호
• /
• pp.146-151
• /
• 2013

With the development of mobile internet, requirements of positioning accuracy for the LBS (Location Based Service) are becoming more and more higher. The LBS is based on the position of each mobile device. So, it requires a proper acquisition of accurate user's indoor position. Thus indoor positioning technology and its accuracy is crucial for various LBS. In general, RSSI (Received Signal Strength Indicator) measurements are used to obtain the position information of mobile unit under WLAN environment. However, indoor positioning error increases as multiple AP's configurations are becoming more complex. To overcome this problem, an enhanced indoor localization method by AP (Access Point) selection criteria adopting DOP (Dilution of Precision) is proposed.

• 제어로봇시스템학회논문지
• /
• 제14권10호
• /
• pp.969-976
• /
• 2008

A position control problem is studied for UEV(Unmanned Expedition Vehicles), which is to follow pre-determined paths via fixed way-points. Hybrid control systems are used for position control of UEV depending on the operating condition. Speed control consists of three controllers: PID control, adaptive PI control, and neural network. Heading control consists of two controllers, PID and adaptive PID control. The controllers are selected based on the changes of road conditions. We suggest an adaptive PI control algorithm for speed control and an transition management algorithm among the controllers. The algorithm adapts the road conditions and variation of vehicle dynamical characteristics and selects a suitable controller.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.252-255
• /
• 2003

In this paper, we suggest the method for a service robot to move safely from an initial position to n goal position in the wide environment like a building. There is a problem using odometry encoder sensor to estimate the position of n mobile robot in the wide environment like a building. Because of the phenomenon of wheel's slipping, a encoder sensor has the accumulated error of n sensor measurement as time. Therefore the error must be compensated with using other sensor. A vision sensor is used to compensate the position of a mobile robot as using the regularly attached light's panel on a building's ceiling. The method to create global path planning for a mobile robot model a building's map as a graph data type. Consequently, we can apply floyd's shortest path algorithm to find the path planning. The effectiveness of the method is verified through simulations and experiments.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 추계학술대회 논문집
• /
• pp.301-305
• /
• 2001

Conventionally, robot control algorithms are divided into two stages, namely, path or trajectory planning and path tracking(or path control). This division has been adopted mainly as a means of alleviating difficulties in dealing with complex, complex, coupled manipulator dynamics. The minimum-time manipulator control problem is solved for the case when the path is specified and the actuator torque limitations are known. In path planning, DP is applied to applied to find the shortest path form initial position to final position with the assumptions that there is no obstacle and that each path is straight line. In path control, the phase plane technique is applied to the minimum-time control with the assumptions that the bound on each actuator torque is a function of joint position and velocity or constant. This algorithm can be used for any manipulator that has rigid link, known dynamics equations of motion, and joint angles that can be determined at a given position on the path.

• Journal of Electrical Engineering and Technology
• /
• 제13권6호
• /
• pp.2376-2384
• /
• 2018

To reduce the size and cost of motor driving systems, several methods for driving multiple parallel-connected motors with a single inverter have been proposed. However, dual PMSMs driven by a single inverter, unlike induction motors, have a problem with instability due to system resonance caused by disturbances such as load imbalance and tolerances between two motors. To drive dual SPMSMs fed by a single inverter, this paper proposes an active damping algorithm to effectively suppress resonance by using one-sided sensorless speed control and position difference estimation. By deriving rotor position difference from d-q current differences between two motors, the proposed method is affected less by position difference estimation errors and is simpler than dual sensorless position estimation.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.1080-1084
• /
• 2007

Many different methods for sound source localization have been developed. Most of them mainly depend on time delay of arrival (TDOA) or on empirical or analytic head related transfer functions (HRTFs). In real implementation, since the direct path between a source and a sensor is interrupted by obstacles as like a head or body of robot, it has to be considered the number of sensors as well as their positions. Therefore, in this paper, we present the methods, which are included sensor position problem, to localize the sound source with 4 microphones to cover the 3D space. Those are modified two-step TDOA methods. Our conclusion is that the different method has to be applied in case to be different microphone position on real robot platform.

• 전기학회논문지
• /
• 제63권9호
• /
• pp.1245-1247
• /
• 2014

This paper presents a robust position tracking controller for motor-driven flexible joint manipulators using only the motor angle measurement. The control problem is not easy because the link position is hard to estimate in the presence of parameter uncertainties. The proposed controller consists of a feedback linearization controller (FLC) and two proportional-integral observers (PIOs) that estimate both system states including the link position and an equivalent disturbance for compensating the parameter uncertainties. Comparative computer simulations are conducted to demonstrate the effectiveness of the proposed control algorithm.