• Journal of the Optical Society of Korea
• /
• v.19 no.4
• /
• pp.340-345
• /
• 2015

The Internet traffic is forecasted to grow at a compound annual rate of 21 % from 2013 to 2018, according to surveys carried out by Cisco [1]. Network resources are significantly over-provisioned in today's networks, and it is quite common to see link utilization in the 30-40% range [2]. Additionally, the multi-media services have widely divergent bandwidth and Quality of Experience (QoE) requirements. Unfortunately, the huge transmission capacity will increase the power consumption of network equipment [3]. Applications Coordinated with Transport, Internet Protocol and Optical Networks (ACTION) [4] has been proposed to realize a multi-QoE, application-centric, and highly energy-efficient network that leverages flexible elastic optical network technologies [5-7]. This paper provides key network technologies for realizing the ACTION, which are a virtual optical slice core network and a Time Division Multiplexing (TDM)-based deterministic active optical access network.

• Journal of Korea Multimedia Society
• /
• v.8 no.6
• /
• pp.844-855
• /
• 2005

This paper proposes a human action recognition scheme for Intelligent Web Browser. Based on the principle that a human action can be defined as a combination of multiple articulation movements, the inference of stochastic grammars is applied to recognize each action. Human actions in 3 dimensional (3D) world coordinate are measured, quantized and made into two sets of 4-chain-code for xy and zy projection planes, consequently they are appropriate for applying the stochastic grammar inference method. We confirm this method by experiments, that various physical actions can be classified correctly against a set of real world 3D temporal data. The result revealed a comparatively successful achievement of $93.8\%$ recognition rate through the experiments of 8 movements of human head and $84.9\%$ recognition rate of 60 movements of human upper body. We expect that this scheme can be used for human-machine interaction commands in a web browser.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.552-554
• /
• 1998

This paper deals with the gait generation of IWR-III using a kick action to have a walking pattern like human. For this, trajectory planning with the consideration of kick action is done in each walking step, and the coordinate transformation is done for simplifying the kinematics. Balancing motion is analyzed by FDM during the walking, By combining 4-types of pre-defined steps, multi-step walking is done. Using numerical simulator, dynamic analysis, ZMP analysis and system stability is confirmed. Walking motion is visualized by 3D- graphic simulator. As a result, trunk ahead motion effect and impactless smooth walking is implemented by experiment. Finally walking with kick action is implemented the IWR-III system.

• Proceedings of the Korean Institute of Interior Design Conference
• /
• 2005.10a
• /
• pp.39-43
• /
• 2005

Here this Study is purposed importance of Public area and attend to planning of Admission Space so foster the growth of public area in super high - rise building apartment at present that, for contribution so Consider about Admission Space planning purpose of a Major building in super high - rise building apartment. The process of this Study 1) Consider Indispensable action and Choice action in theory of 'Jan Gehl'. 2) Express one's condolence about two housing development of super high - rise Apartment in Busan City. 3) Analysis by theory of 'Jan Gehl'. about two housing development of super high - rise Apartment in Busan City. 4) I have a result as being Comparison two housing development

• ETRI Journal
• /
• v.43 no.6
• /
• pp.1004-1012
• /
• 2021

The collaboration productively interacting between multi-agents has become an emerging issue in real-world applications. In reinforcement learning, multi-agent environments present challenges beyond tractable issues in single-agent settings. This collaborative environment has the following highly complex attributes: sparse rewards for task completion, limited communications between each other, and only partial observations. In particular, adjustments in an agent's action policy result in a nonstationary environment from the other agent's perspective, which causes high variance in the learned policies and prevents the direct use of reinforcement learning approaches. Unexpected social loafing caused by high dispersion makes it difficult for all agents to succeed in collaborative tasks. Therefore, we address a paradox caused by the social loafing to significantly reduce total returns after a certain timestep of multi-agent reinforcement learning. We further demonstrate that the collaborative paradox in multi-agent environments can be avoided by our proposed effective early stop method leveraging a metric for social loafing.

• Journal of the Korea Safety Management & Science
• /
• v.6 no.2
• /
• pp.177-186
• /
• 2004

Decision analysis has becomes an important technique for decision making in the face of uncertainty. It is characterized by enumerating all the available courses of action, identifying the payoffs for all possible outcomes, and quantifying the subjective probabilities for the all possible random events. When the data are available, decision analysis becomes a powerful tool for determining an optimal course of action. We study the multi-attribute decision making in a compensatory models. In this paper, we use the entropy methods in weights calculating. For the purpose of making optimal decision, the data of five different car models are used. For computing, we used Visual Numerica Version 1.0 software package.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 1999.06b
• /
• pp.185-194
• /
• 1999

This paper suggests the new techmology to control metal flow in order to reduce the number of preforming and machining for the cold forged product with complex geometry. This technology can be summarized the complex forming, which consists of bulk forming and sheet forming, and multi-action forging, which be preformed double action dies. To analyze the process, finite element simulation has been performed. The proposed technology is applied to hub which is part of air conditioner clutch. According to the result of this study, the relative velocity of mandrel and punch is primary process variable.

• Steel and Composite Structures
• /
• v.11 no.2
• /
• pp.149-168
• /
• 2011

This paper presents a computational study of column loss scenarios for typical multi-story steel buildings with perimeter moment frames and composite steel-concrete floors. Two prototype buildings (three-story and ten-story) were represented using three-dimensional nonlinear finite element models and explicit dynamic analysis was used to simulate instantaneous loss of a first-story column. Twelve individual column loss scenarios were investigated in the three-story building and four in the ten-story building. This study provides insight into: three-dimensional load redistribution patterns; demands on the steel deck, concrete slab, connections and members; and the impact of framing configuration, building height and column loss location. In the dynamic simulations, demands were least severe for perimeter columns within a moment frame, but the structures also exhibited significant load redistribution for interior column loss scenarios that had no moment connectivity. Composite action was observed to be an important load redistribution mechanism following column loss and the concrete slab and steel deck were subjected to high localized stresses as a result of the composite action. In general, the steel buildings that were evaluated in this study demonstrated appreciable robustness.

• Journal of Intelligence and Information Systems
• /
• v.10 no.3
• /
• pp.91-103
• /
• 2004

Multi-agent systems for semantic web applications require efficient implementation of agent architectures without sacrificing the flexibility and the level of abstraction that agent architectures provide. In this paper, we present an agent system, called VivAce, which is implemented in Java to achieve both high efficiency and the level of abstraction provided by the BDI agent architecture. VivAce (Vivid Agent Computing Environment) has the characteristics of a vivid agent through the BDI agent model. A vivid agent is a software-controlled system whose state comprises the mental components of knowledge, perceptions, tasks, and intentions, and whose behavior is represented by means of action and reaction rules. We first identify the requirements for multi-agent systems and then present the relevant features of VivAce and experimental results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.4 no.3
• /
• pp.296-300
• /
• 2003

A large equilibrium and kinetic data set for multi-component cation exchanges was obtained and tested with mass action law and surface complexation model. The systematic batch equilibrium and column experiments of cation adsorption were conducted for binary, ternary, quaternary, and quinary cation exchanges involving $ H^{+}, Li^ {+}, Na^{+}, NH$_4$^{+}, Mg^{2+} $ on a strongly acidic cation exchange resin IRN 77. The mass action law and surface complexation model were tested against both data set to investigate the consistency of ion selectivity and their predictability for competitive cation exchanges. Surface complexation model provided more accurate predictions for both equilibrium and kinetic experimental data than mass action model.