• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.5
• /
• pp.115-123
• /
• 2015

The incremental learning method performs well in face face tracking. However, it has a drawback in that it is sensitive to the tracking error in the previous frame due to the environmental changes. In this paper, we propose a selective incremental learning method to track a face more reliably under various conditions. The proposed method is robust to illumination variation by using the LBP(Local Binary Pattern) features for each individual frame. We select patches to be used in incremental learning by using Staggered Multi-Scale LBP, which prevents the propagation of tracking errors occurred in the previous frame. The experimental results show that the proposed method improves the face tracking performance on the videos with environmental changes such as illumination variation.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.232-240
• /
• 2009

This paper presents a robust vehicle detection and tracking algorithm for supervision of illegal parking. The proposed algorithm is composed of four parts. First, a vehicle detection algorithm is proposed using the improved codebook object detection algorithm to segment moving vehicles from the input sequence. Second, a preprocessing technique using the geometric characteristics of vehicles is employed to exclude non-vehicle objects. Then, the detected vehicles are tracked by an object tracker which incorporates histogram tracking method with Kalman filter. To make the tracking results more accurate, histogram tracking results are used as measurement data for Kalman filter. Finally, Real Stop Counter (RSC) is introduced for trustworthy and accurate performance of the stopped vehicle detection. Experimental results show that the proposed algorithm can track multiple vehicles simultaneously and detect stopped vehicles successfully in the complicated street environment.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.409-420
• /
• 2018

In this paper, we propose a Motion Vector-based Particle Filter(MVPF) for object tracking on bitstreams and a object tracking system using the MVPF. The MVPF uses motion vectors to both the transition and the observation models of a general particle filter to improve the accuracy while maintaining the number of particles. In the proposed object tracking system, the state of the target object can be predicted using the histogram of motion vectors extracted from the bitstream. In terms of precision, F-measure and IOU(Intersection Of Union), the proposed method is about 30%, 17%, and 17% better on average, respectively, in MPEG test sequences and VOT2013 sequences. Furthermore, When the tracking results are displayed in box form for subjective performance evaluation, the proposed method can track moving objects more robust than the conventional methods in all test sequences.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.2A
• /
• pp.159-166
• /
• 2008

In this paper, we proposed an algorithm to solve the problem that can't maintain hop synchronization using only early-late gate tracking loop due to the drift of geo-stationary satellite in frequency hopping satellite communication system. When the signal is transferred to downlink through DRT(Dehop-Rebop Transponder), the problem with synchronization loss is occurred periodically when using only early-late gate tracking loop, because of energy loss in each side portion of hop due to orbital variation of the satellite. To solve this problem, we have developed Anti-Shrink synchronization tracking algorithm which uses the prediction value of transmission timing and the structure of inner-outer gate instead of early-late gate with the ranging information. Through simulations, we showed that the performance of the Anti-Shrink algorithm is better than that of simple inner-outer energy ratio algorithm and similar to that of conventional early-late tracking loop algorithm with ranging information. No synchronization failure in the proposed algorithm was occurred because of less energy loss and robustness without the ranging information.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.1
• /
• pp.101-108
• /
• 2004

An electro-hydraulic manipulator using hydraulic actuators has many nonlinear elements, and its parameter fluctuations are greater than those of an electrically driven manipulator. So it is relatively difficult to obtain stable control performance. In this report, we applied disturbance estimation and compensation type robust control to all axes in a 3-link electro-hydraulic manipulator. From the results of experiment, it was confirmed that the performance of trajectory tracking and attitude regulating is greatly improved by the disturbance observer, which model is the same for each axis. On the other hand, for the autonomous assembly tasks, it is said that compliance control is one of the most available methods. Therefore we proposed compliance control which is based on the position control by disturbance observer for our manipulator system. To realize more stable contact work, the states in the compliance loop are feedback, where not only displacement but also velocity and acceleration are considered. And we applied this compliance control to Peg-in-Hole insertion task and analyzed mechanical relation between peg and hole. Also we proposed new method of shifting the position of end-effector periodically for the purpose of smooth insertion. As a result of using this method, it is experimentally confirmed that Peg-in-Hole insertion task with a clearance of 0.05［mm］can be achieved.

• International Journal of Automotive Technology
• /
• v.7 no.5
• /
• pp.609-618
• /
• 2006

This paper presents the lateral control of an autonomous vehicle by using a look-ahead sensing system. In look-ahead sensing by an absolute positioning system, a reference lane, constructed by straight lanes or circular lanes, was switched by a segment switching algorithm. To cope with sensor noise and modeling uncertainty, a robust LMI-based $H_{\infty}$ lateral controller was designed by the feedback of lateral offset and yaw angle error at the vehicle look-ahead. In order to verify the safety and the performance of lateral control, a scaled-down vehicle was developed and the location of the vehicle was detected by using an ultrasonic local positioning system. In the mechatronic scaled-down vehicle, the lateral model and parameters are verified and estimated by a J-turn test. For the lane change and reference lane tracking, the lateral controllers are used experimentally. The experimental results show that the $H_{\infty}$ controller is robust and has better performance compared with look-down sensing.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.1
• /
• pp.47-53
• /
• 2004

An electro-hydraulic manipulator using hydraulic actuators has many nonlinear elements, and its parameter fluctuations are greater than those of an electrically driven manipulator. So it is quite difficult to obtain stable control performance. We have applied a disturbance estimation and compensation type robust control to all the axes in a 6-link electro-hydraulic manipulator. It was confirmed that the performance of trajectory tracking and attitude regulating was greatly improved by the disturbance observer. For autonomous assembly tasks, it is said that compliance control is one of the most popular methods in contact task. We have proposed a compliance control based on the position control by a disturbance observer for our manipulator system. To realize more stable contact work, the states in the compliance loop are feedbacked, where not only displacement but also the velocity and acceleration are considered. We have also applied this compliance control to the Peg-in-Hole insertion task and proposed new methods of (1)rotating of the end-effector periodically in order to reduce the friction force, (2)random searching for the center of a hole and (3)trajectory modification to reduce the impact force. As a result of these new methods, it could be experimentally confirmed that the Peg-in-Hole insertion task with a clearance of 0.007 [mm] could be achieved.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.1
• /
• pp.45-55
• /
• 1998

In this paper, a robust path tracker and depth controller of Autonomous Underwater Vehicle based on sliding mode control is presented. We have also designed augmented equivalent control inputs by analyzing the sliding mode with the reaching mode. This can enhance the reaching rate, and improve chattering problems, that is, noise caused by the control plane actuator of the vehicle, which is one of the problems that occur when sliding mode control is used. Also to resolve the steady state error generated in the path tracker under current effect, a modified sliding plane is constructed. Also a redesigned sliding plane and control input using transformation matrix is proposed to do easy design of MIMO depth controller. For state variables that cannot be measured directly, reduced order sliding mode control is used to design an observer. The performance of designed path tracker and depth controller is investigated by computer simulation. The results show that the proposed control system has robust performance to parameter variation, modelling error and disturbance.

• Journal of the Korean Society of Industry Convergence
• /
• v.19 no.2
• /
• pp.81-87
• /
• 2016

This paper presents a new approach to control of stable motion of single wheel driving robot system of a pitch that is controlled by an in-wheel motor and a roll that is controlled by a reaction wheel. This robot doesn'thave any actuator for a yaw axis control, which makes the derivation of the dynamics relatively simple. The Lagrange equations was applied to derive the dynamic equations of the one wheel driving robot to implement the dynamic speed control of the mobile robot. To achieve the real time speed control of the unicycle robot, the sliding mode control and optical regulator are utilized to prove the reliability while maintaining the desired speed tracking performance. In the roll controller, the sigmoid-function based robust controller has been adopted to reduce the vibration by the situation function. The optimal controller has been implemented for the pitch control to drive the unicycle robot to follow the desired velocity trajectory in real time using the state variables of pitch angle, angular velocity, angle and angular velocity of the driving wheel. The control performance of the control systems from a single dynamic model has been illustrated by the real experiments.

• International Journal of Automotive Technology
• /
• v.8 no.5
• /
• pp.555-562
• /
• 2007

The Variable Valve Timing(VVT) system for high performance is a key technology used in newly developed engines. The system realizes higher torque, better fuel economy, and lower emissions by allowing an additional degree of freedom in valve timing during engine operation. In this study, a model-based control method is proposed to enable a fast and precise VVT control system that is robust with respect to manufacturing tolerances and aging. The VVT system is modeled by a third-order nonlinear state equation intended to account for nonlinearities of the system. Based on the model, a controller is designed for position control of the VVT system. The sliding mode theory is applied to controller design to overcome model uncertainties and unknown disturbances. The experimental results suggest that the proposed sliding mode controller is capable of improving tracking performance. In addition, the sliding mode controller is robust to battery voltage disturbance.