• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2796-2798
• /
• 2005

This Paper presents a method that a mobile robot can track persons in complex environment using particle filters. The topic of person following using mobile robot is researched in many different areas. The main problems of following a person are real time constraint, motion change of person during the tracking and occlusion with other objects. We present appearance adaptive models in a particle filter to realize robust visual tracking algorithm. Adaptive appearance model can handle occlusion with other people while target is moving.

• 한국전산유체공학회:학술대회논문집
• /
• 2007.04a
• /
• pp.53-57
• /
• 2007

The quality of chaotic mixing in square cavity flow was studied numerically by CFD simulation and particle tracking technique. The chaotic mixing was generated by using time-periodic electro-osmotic flow. Finite Volume Method (FVM) was employed to get the stretching and folding field in cavity domain. With adjusting the initial condition of concentration distribution, the best values of modulation period and Peclet number which gave us good mixing performance was determined precisely. From $Poicar{\acute{e}}section$and Lyapunov exponents for characteristic trajectories we find that mixing performance also depends on modulation period. The higher value of modulation period, the better mixing performance wag achieved in this case. Furthermore, the results for tracking particle trajectories were also compared between using of Bilinear Interpolation and Higher-order scheme. The values of modulation period for obtaining best mixing effect were matched between using FVM and particle tracking techniques.

• Journal of Mechanical Science and Technology
• /
• v.15 no.1
• /
• pp.108-116
• /
• 2001

The in-cylinder flow field of gasoline engine comprises unsteady compressible turbulent flows caused by the intake port, combustion chamber geometry. Thus, the quantitative analysis of the in-cylinder flow characteristics plays an important role in the improvement of engine performances and the reduction of exhaust emission. In order to obtain the quantitative analysis of the in-cylinder gas flows for a gasoline engine, the single-frame particle tracking velocimetry was developed, which is designed to measure 2-dimensional gas flow field. In this paper, influences of the swirl and tumble intensifying valves on the in-cylinder flow characteristics under the various intake flow conditions were investigated by using this PTV method. Based on the results of experiment, the generation process of swirl and tumble flow in a cylinder during intake stroke was clarified. Its effect on the tumble ratio at the end of compression stroke was also investigated.

• Nuclear Engineering and Technology
• /
• v.38 no.8
• /
• pp.763-778
• /
• 2006

Injection of microbubbles within the turbulent boundary layer has been investigated for several years as a method to achieve drag reduction. However, the physical mechanism of this phenomenon is not yet fully understood. Experiments in a channel flow for single phase (water) and two phase (water and microbubbles) flows with various void fraction values are studied for a Reynolds number of 5128 based on the half height of the channel and bulk velocity. The state-of-the art Particle Tracking Velocimetry (PTV) measurement technique is used to measure the instantaneous full-field velocity components. Comparisons between turbulent statistical quantities with various values of local void fraction are presented to elucidate the influence of the microbubbles presence within the boundary layer. A decrease in the Reynolds stress distribution and turbulence production is obtained with the increase of microbubble concentration. The results obtained indicate a decorrelation of the streamwise and normal fluctuating velocities when microbubbles are injected within the boundary layer.

• Journal of Internet Computing and Services
• /
• v.17 no.3
• /
• pp.25-32
• /
• 2016

In this paper, we propose a method that is tracking an object in real time using particle filter and the observation model with edge. First of all, the proposed method defines the object to be tracked in the initial frame. Then, it generates the edge observation model for the object to be tracked and a set of particles. It calculates the weight by comparing the average of the middle distance in eight-way of particle filter edge model with that in edge observation model, and then updates the weight with the calculated value. After resampling particles using the updated weights, it estimates the current location of the tracked object. Finally, this paper demonstrates the performance of the stable tracking through comparison with the existing method by using a number of experimental data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.7
• /
• pp.638-643
• /
• 2012

This paper presents an intelligent reduction method for external noise. The main idea comes from PSO-FCM (Particle Swam Optimization Fused fuzzy C-Means) clustering. The data of the target is transformed from the antenna coordinates to the vessel one and to the system coordinates. In the conversion, the overall noises hinder observer to get the exact position and velocity of the maneuvering target. While the filter is used for tracking system, unexpected acceleration becomes the main factor which makes the uncertainty. In this paper, the tracking efficiency is improved with the PSO-FCM and the compensation methodology. The acceleration is approximated from the external noise splitted by the proposed clustering method. After extracting the approximated acceleration, the rest in the noise is filtered by the filter and the compensation is added to after that. Proposed tracking method is applicable to the linear model and nonlinear one together. Also, it can do to the on-line system. Finally, some examples are provided to examine the reliability of the proposed method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.586-591
• /
• 2009

In this paper, we propose the mobile object tracking algorithm based on the feature vector using particle filter. To do this, first, we detect the movement area of mobile object by using RGB color model and extract the feature vectors of the input image by using the KLT-algorithm. And then, we get the first feature vectors by matching extracted feature vectors to the detected movement area. Second, we detect new movement area of the mobile objects by using RGB and HSI color model, and get the new feature vectors by applying the new feature vectors to the snake algorithm. And then, we find the second feature vectors by applying the second feature vectors to new movement area. So, we design the mobile object tracking algorithm by applying the second feature vectors to particle filter. Finally, we validate the applicability of the proposed method through the experience in a complex environment.

• Nuclear Engineering and Technology
• /
• v.53 no.7
• /
• pp.2334-2340
• /
• 2021

Radioactive particle tracking (RPT) is a minimally invasive nuclear technique that tracks a radioactive particle inside a volume of interest by means of a mathematical location algorithm. During the past decades, many algorithms have been developed including ones based on artificial intelligence techniques. In this study, RPT technique is applied in a simulated test section that employs a simplified mixer filled with concrete, six scintillator detectors and a¹³⁷Cs radioactive particle emitting gamma rays of 662 keV. The test section was developed using MCNPX code, which is a mathematical code based on Monte Carlo simulation, and 3516 different radioactive particle positions (x,y,z) were simulated. Novelty of this paper is the use of a location algorithm based on a deep learning model, more specifically a 6-layers deep rectifier neural network (DRNN), in which hyperparameters were defined using a Bayesian optimization method. DRNN is a type of deep feedforward neural network that substitutes the usual sigmoid based activation functions, traditionally used in vanilla Multilayer Perceptron Networks, for rectified activation functions. Results show the great accuracy of the DRNN in a RPT tracking system. Root mean squared error for x, y and coordinates of the radioactive particle is, respectively, 0.03064, 0.02523 and 0.07653.

• Nuclear Engineering and Technology
• /
• v.42 no.2
• /
• pp.211-218
• /
• 2010

An analysis model on debris transport in the containment floor of pressurized water reactors is developed in which the flow field is calculated by Eulerian conservation equations of mass and momentum and the debris particles are traced by Lagrange equations of motion using the pre-determined flow field data. For the flow field calculation, two-dimensional Shallow Water Equations derived from Navier Stokes equations are solved using the Finite Volume Method, and the Harten-Lax-van Leer scheme is used for accuracy to capture the dry-to-wet interface. For the debris tracing, a simplified two-dimensional Lagrangian particle tracking model including drag force is developed. Advanced schemes to find the positions of particles over the containment floor and to determine the position of particles reflected from the solid wall are implemented. The present model is applied to calculate the transport fraction to the Hold-up Volume Tank in Advanced Power Reactors 1400. By the present model, the debris transport fraction is predicted, and the effect of particle density and particle size on transport is investigated.

• The KIPS Transactions:PartB
• /
• v.17B no.1
• /
• pp.37-46
• /
• 2010

In this paper we have presented a two dimensional model based tracking system using improved chamfer matching. Conventional chamfer matching could not calculate similarity well between the object and image when there is very cluttered background. Then we have improved chamfer matching to calculate similarity well even in very cluttered background with edge and corner feature points. Improved chamfer matching is used as likelihood function of particle filter which tracks the geometric object. Geometric model which uses edge and corner feature points, is a discriminant descriptor in color changes. Particle Filter is more non-linear tracking system than Kalman Filter. Then the presented method uses geometric model, particle filter and improved chamfer matching for tracking object in complex environment. In experimental result, the robustness of our system is proved by comparing other methods.