• ETRI Journal
• /
• 제30권2호
• /
• pp.338-340
• /
• 2008

In this letter, we propose a novel approach to detecting and tracking apartment buildings for the development of a video-based navigation system that provides augmented reality representation of guidance information on live video sequences. For this, we propose a building detector and tracker. The detector is based on the AdaBoost classifier followed by hierarchical clustering. The classifier uses modified Haar-like features as the primitives. The tracker is a motion-adjusted tracker based on pyramid implementation of the Lukas-Kanade tracker, which periodically confirms and consistently adjusts the tracking region. Experiments show that the proposed approach yields robust and reliable results and is far superior to conventional approaches.

• Journal of Mechanical Science and Technology
• /
• 제17권1호
• /
• pp.32-39
• /
• 2003

In this study we examined connections between vertebral motion and patterns of muscle activation during voluntary head tracking movements. A Rhesus (Maraca mulatta) monkey was trained to produce sinusoidal tracking movements of the head in the sagittal plane while seated. Radio-opaque markers were placed in the cervical vertebrae, and intramuscular patch electrodes were implanted to record from eight neck muscles. Videofluoroscopic images of cervical vertebral motion, and EMG (electromyographic) responses were simultaneously re-corded. Experimental results demonstrated that head and vertebrae moved synchronously and that motion occurred primarily at skull-C$_1$, C$\_$6/-C$\_$7/ and Csub 7/-C$_1$. Our findings illustrate that although the biomechanical constraints of each species may limit the number of solutions available, it is the task requirements that appear to govern CNS (central nervous system) selection of movement behaviors.

• 한국정밀공학회지
• /
• 제26권12호
• /
• pp.69-77
• /
• 2009

In this study an ANFIS-based trajectory tracking motion control algorithm is proposed for autonomous garage and parallel parking of a model car. The ANFIS controller is trained off-line using data set which obtained by Mandani fuzzy inference system and thereby the processing time decreases almost in half. The controller with a steering delay compensator is tuned through simulations performed under MATLAB/Simulink environment. Experiments are carried out with the model car for garage and parallel parking. The experimental results show that the trajectory tracking performance is satisfactory under various initial and road conditions

• International Journal of Control, Automation, and Systems
• /
• 제3권1호
• /
• pp.87-99
• /
• 2005

We present in this paper the stabilization (tracking) with motion planning of the six independent configurations of a mini unmanned areal vehicle equipped with four streamlined rotors. Naturally, the yaw-dynamic can be stabilized without difficulties and independently of other motions. The remaining dynamics are linearly approximated around a small roll and pitch angles. It will be shown that the system presents a flat output that is likely to be useful in the motion generation problem. The tracking feedback controller is based on receding horizon point to point steering. The resulting controller involves the lift (collective) time derivative for what flatness and feedback linearization are used. Simulation tests are performed to progress in a region with approximatively ten-meter-buildings.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.1574-1579
• /
• 2004

A simultaneous measurement system that can analyze the flow-structure interactions(FSI) has been constructed and analyses on the flow field and the motion field of a floating cylinder was made. The three-dimensional vector fields around the cylinder are measured by 3D-PTV technique while the motion of the cylinder forced by the flow field is measured simultaneously with a newly developed motion tracking algorithm(bidirectional tracking algorithm). The cylinder is pendant in the working fluid of a water channel and the surface of the working fluid is forced sinusoidal to make the cylinder bounced. The interaction between the flow fields and the cylinder motion is examined quantitatively.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.1816-1819
• /
• 1997

Using the vision system, robotic tasks in unstructured environments can be accompished, which reduces greatly the cost and steup time for the robotic system to fit to he well-defined and structured working environments. This paper proposes a dynamic control scheme for robot manipulator with eye-in-hand camera configuration. To perfom the tasks defined in the image plane, the camera motion Jacobian (image Jacobian) matrix is used to transform the camera motion to the objection position change. In addition, the dynamic learning controller is designed to improve the tracking performance of robotic system. the proposed control scheme is implemented for tasks of tracking moving objects and shown to outperform the conventional visual servo system in convergence and robustness to parameter uncertainty, disturbances, low sampling rate, etc.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2013년도 추계학술대회
• /
• pp.1015-1017
• /
• 2013

물체의 움직임에 관한 추적방법은 여러 가지 문제점을 갖고 있다. 물체의 움직임에 관한 추적방법은 물제의 장면, 비 강체 물체의 구조, 물체와 물체 및 물체의 장면 폐색 및 카메라의 움직임과 모두 움직이는 물체의 패턴변화에 의해 결정되기 때문이다. 추적방법은 일반적으로 매 프레임의 위치나 물체의 형상을 필요로 하는 높은 수준의 응용프로그램이나 시스템 내에서 처리된다. 본 논문에서는 확장 칼만 필터(EKF)에 따라 물체의 활성 시각 추적 물체 잠금 시스템을 실행하고, 실행된 데이터를 바탕으로 분석하여 도입된 단일 카메라 추적 시스템 알고리즘에 2대의 카메라와 각각의 비전에 따라 물체 추적 시스템을 설명하고, 물체의 상태를 파악하여 각 카메라에서 움직임에 관한 추적이 실행된 후 개별 트랙에 최종 시스템 물체의 움직임 트랙과 결합하여 사용되는 추적시스템에 대해 연구하였다.

• 한국멀티미디어학회논문지
• /
• 제23권10호
• /
• pp.1236-1249
• /
• 2020

In a most recent object tracking research work, applying Convolutional Neural Network and Recurrent Neural Network-based strategies become relevant for resolving the noticeable challenges in it, like, occlusion, motion, object, and camera viewpoint variations, changing several targets, lighting variations. In this paper, the LSTM Network-based Tracking association method has proposed where the technique capable of real-time multi-object tracking by creating one of the useful LSTM networks that associated with tracking, which supports the long term tracking along with solving challenges. The LSTM network is a different neural network defined in Keras as a sequence of layers, where the Sequential classes would be a container for these layers. This purposing network structure builds with the integration of tracking association on Keras neural-network library. The tracking process has been associated with the LSTM Network feature learning output and obtained outstanding real-time detection and tracking performance. In this work, the main focus was learning trackable objects locations, appearance, and motion details, then predicting the feature location of objects on boxes according to their initial position. The performance of the joint object tracking system has shown that the LSTM network is more powerful and capable of working on a real-time multi-object tracking process.

• 한국기계가공학회지
• /
• 제20권11호
• /
• pp.43-51
• /
• 2021

A robust tracking controller design was developed for a rotary motion control system. The friction force versus the angular velocity was measured and modeled as a combination of linear and nonlinear components. By adding a model-based friction compensator to a nominal proportional-integral-derivative controller, it was possible to build a simulated control system model that agreed well with the experimental results. A zero-phase error tracking controller was selected as the feedforward tracking controller and implemented based on the estimated closed-loop transfer function. To provide robustness against external disturbances and modeling uncertainties, a disturbance observer was added in the position feedback loop. The performance improvement of the overall tracking controller structure was verified through simulations and experiments.

• 한국멀티미디어학회논문지
• /
• 제10권6호
• /
• pp.716-725
• /
• 2007

In this paper we describe a motion tracking algorithm for 3D human animation using stereo vision system. This allows us to extract the motion data of the end effectors of human body by following the movement through segmentation process in HIS or RGB color model, and then blob analysis is used to detect robust shape. When two hands or two foots are crossed at any position and become disjointed, an adaptive algorithm is presented to recognize whether it is left or right one. And the real motion is the 3-D coordinate motion. A mono image data is a data of 2D coordinate. This data doesn't acquire distance from a camera. By stereo vision like human vision, we can acquire a data of 3D motion such as left, right motion from bottom and distance of objects from camera. This requests a depth value including x axis and y axis coordinate in mono image for transforming 3D coordinate. This depth value(z axis) is calculated by disparity of stereo vision by using only end-effectors of images. The position of the inner joints is calculated and 3D character can be visualized using inverse kinematics.