• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.06a
• /
• pp.381-384
• /
• 2011

본 논문은 사용자 및 동적환경의 변화를 파악하고 분석된 정보를 바탕으로 최적화된 경로를 제공하기 위한 시스템을 멀티에이전트를 이용하여 해결하고자 하였다. 멀티에이전트를 통해 설정된 목표를 찾아가는 먹이추적 문제에 적용하였고 현실 세계와 흡사한 무한 공간 환경에서 알고리즘의 성능을 실험하였다. 적용된 환경의 모델은 순환구조(circular)형 격자 공간이라는 새로운 실험 공간으로 방향 벡터 함수 알고리즘을 통해 새롭게 멀티에이전트의 목표를 획득하기 위한 해법이다. 기존의 연구와 비교하여 먹이의 효율적 포획, 에이전트간의 충돌문제 해결에 대한 새로운 해법을 제시할 수 있었다.

• Communications for Statistical Applications and Methods
• /
• v.28 no.1
• /
• pp.21-37
• /
• 2021

There is a direct connection between linear discriminant analysis (LDA) and linear regression since the direction vector of the LDA can be obtained by the least square estimation. The connection motivates the penalized LDA when the model is high-dimensional where the number of predictive variables is larger than the sample size. In this paper, we study the penalized LDA for a class of penalties, called the moderately clipped LASSO (MCL), which interpolates between the least absolute shrinkage and selection operator (LASSO) and minimax concave penalty. We prove that the MCL penalized LDA correctly identifies the sparsity of the Bayes direction vector with probability tending to one, which is supported by better finite sample performance than LASSO based on concrete numerical studies.

• Journal of the Korean Society for Precision Engineering
• /
• v.10 no.4
• /
• pp.216-226
• /
• 1993

Three dimensional cutting is considered as an equivalent orthogonal cutting through the plane containing both the cutting velocity vector and the chip flow velocity vector in dynamic cutting process. An analytical expression of dynamic cutting force is obtained from the cutting parameters determined by the static cutting. Particular attention is paid to the energy supplied to the vibratory system of cutting tool with two degree of freedom. In this approach, the phase lag of the horizontal vibration of the tool behind the vertical vibration and the direction angle of the fluctuating cutting force is considered in point of stability limits. Chatter vibration can be effectively suppressed by relatively increasing the spring constant and the damping coefficient of the cutting system in the vertical cutting force direction. A good agreement is found between the stability limits predicted by theoretical value and experimental results.

• Steel and Composite Structures
• /
• v.46 no.6
• /
• pp.719-729
• /
• 2023

This paper studies electro-magneto-mechanical bending studying of the cylindrical panels based on shear deformation theory. The cylindrical panel is constrained with two simply-supported edges at longitudinal direction and two clamped boundary conditions at circumferential direction. The governing equations are derived based on the principle of virtual work in cylindrical coordinate system. Levy-type solution of the governing equations is derived to reduce two dimensional PDEs to a 2D ODEs. The reduced ordinary differential equation is solved using the Eigen-value Eigen-vector method for the clamped-clamped boundary condition. The electro-magneto-mechanical bending results are obtained to show that every displacement, rotation and electromagnetic potentials how change with changes of initial electromagnetic potentials and mechanical loads along longitudinal and circumferential directions.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.518-528
• /
• 2014

In this paper, we propose a method to recognize the action direction of human by developing 4D space-time (4D-ST, [x,y,z,t]) features. For this, we propose 4D space-time interest points (4D-STIPs, [x,y,z,t]) which are extracted using 3D space (3D-S, [x,y,z]) volumes reconstructed from images of a finite number of different views. Since the proposed features are constructed using volumetric information, the features for arbitrary 2D space (2D-S, [x,y]) viewpoint can be generated by projecting the 3D-S volumes and 4D-STIPs on corresponding image planes in training step. We can recognize the directions of actors in the test video since our training sets, which are projections of 3D-S volumes and 4D-STIPs to various image planes, contain the direction information. The process for recognizing action direction is divided into two steps, firstly we recognize the class of actions and then recognize the action direction using direction information. For the action and direction of action recognition, with the projected 3D-S volumes and 4D-STIPs we construct motion history images (MHIs) and non-motion history images (NMHIs) which encode the moving and non-moving parts of an action respectively. For the action recognition, features are trained by support vector data description (SVDD) according to the action class and recognized by support vector domain density description (SVDDD). For the action direction recognition after recognizing actions, each actions are trained using SVDD according to the direction class and then recognized by SVDDD. In experiments, we train the models using 3D-S volumes from INRIA Xmas Motion Acquisition Sequences (IXMAS) dataset and recognize action direction by constructing a new SNU dataset made for evaluating the action direction recognition.

• Journal of information and communication convergence engineering
• /
• v.15 no.3
• /
• pp.137-142
• /
• 2017

A general linearly constrained narrowband adaptive array is examined in the eigenvector space. The optimum weight vector in the eigenvector space is shown to have the same performance as in the standard coordinate system, except that the input signal correlation matrix and look direction steering vector are replaced with the eigenvalue matrix and transformed steering vector. It is observed that the variation in gain factor results in the variation in the distance between the constraint plane and the origin in the translated weight vector space such that the increase in gain factor decreased the distance from the constraint plane to the origin, thus affecting the nulling performance. Simulation results showed that the general linearly constrained adaptive array performed better at an optimal gain factor compared with the conventional linearly constrained adaptive array in a coherent signal environment and the former showed similar performance as the latter in a noncoherent signal environment.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.19 no.4
• /
• pp.44-50
• /
• 2011

This paper presents the algorithm for attitude determination of an aircraft using binary image. An image feature vector, which is invariant to translation, scale and rotation, is constructed to capture the functional relations between the feature vector and the corresponding aircraft attitude. An iterated least squares method is suggested for estimating the attitude of given aircraft using the constructed feature vector library. Simulation results show that the proposed algorithm yields good estimates of aircraft attitude in most viewing range, although a relatively large error occurs in some limited viewing direction.

• Bulletin of the Korean Mathematical Society
• /
• v.51 no.2
• /
• pp.531-538
• /
• 2014

Concentric hyperspheres in the n-dimensional Euclidean space $\mathbb{R}^n$ are the level hypersurfaces of a radial function f : $\mathbb{R}^n{\rightarrow}\mathbb{R}$. The magnitude $||{\nabla}f||$ of the gradient of such a radial function f : $\mathbb{R}^n{\rightarrow}\mathbb{R}$ is a function of the function f. We are interested in the converse problem. As a result, we show that if the magnitude of the gradient of a function f : $\mathbb{R}^n{\rightarrow}\mathbb{R}$ with isolated critical points is a function of f itself, then f is either a radial function or a function of a linear function. That is, the level hypersurfaces are either concentric hyperspheres or parallel hyperplanes. As a corollary, we see that if the magnitude of a conservative vector field with isolated singularities on $\mathbb{R}^n$ is a function of its scalar potential, then either it is a central vector field or it has constant direction.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.4
• /
• pp.379-384
• /
• 2009

A human gaze detection and tracing method is importantly required for HMI(Human-Machine-Interface) like a Human-Serving robot. This paper proposed a novel three-dimension (3D) human gaze estimation method by using a face recognition, an orientation estimation and SVMs (Support Vector Machines). 2,400 images with the pan orientation range of $-90^{\circ}{\sim}90^{\circ}$ and tilt range of $-40^{\circ}{\sim}70^{\circ}$ with intervals unit of $10^{\circ}$ were used. A stereo camera was used to obtain the global coordinate of the center point between eyes and Gabor filter banks of horizontal and vertical orientation with 4 scales were used to extract the facial features. The experiment result shows that the error rate of proposed method is much improved than Liddell's.

• Journal of the Korean Mathematical Society
• /
• v.31 no.3
• /
• pp.339-352
• /
• 1994

A sumbanifold M of a quaternionic Kaehlerian manifold $\tilde{M}^m$ of real dimension 4m is called a generic submanifold if the normal space N(M) of M is always mapped into the tangent space T(M) under the action of the quaternionic Kaehlerian structure tensors of the ambient manifold at the same time.The purpose of the present paper is to study generic submanifold of quaternionic Kaehlerian manifold of constant Q-sectional curvature with nonvanishing parallel mean curvature vector. In section 1, we state general formulas on generic submanifolds of a quaternionic Kaehlerian manifold of constant Q-sectional curvature. Section 2 is devoted to the study generic submanifolds with nonvanishing parallel mean curvature vector and compute the restricted Laplacian for the second fundamental form in the direction of the mean curvature vector. As applications of those results, in section 3, we prove our main theorems. In this paper, the dimension of a manifold will always indicate its real dimension.