• Korean Journal of Computational Design and Engineering
• /
• v.15 no.4
• /
• pp.306-313
• /
• 2010

In this paper, an acceleration based passenger evacuation simulation is performed. In order to describe a passenger‘s behavior in an evacuation situation, a passenger is modeled as a rigid body which translates in the horizontal plane and rotates along the vertical axis. The position and rotation angle of a passenger are calculated by solving the dynamic equations of motions at each time step. The destination force, the contact force, and the group force are considered as external forces and the moments due to each force are also considered. With the passenger model proposed in this paper, the test problems in International Maritime Organization, Maritime Safety Committee/Circulation 1238(IMO MSC/Circ.1238) are implemented and the effects of passenger rotation on the evacuation time are confirmed.

• Journal of Korea Multimedia Society
• /
• v.10 no.6
• /
• pp.770-778
• /
• 2007

Fish schooling or group moving in cyber underwater is a part of beautiful and familiar ecosystem. It is not so easy to present the behavior of fish crowd naturally as a computer animation. Thanks to development of computer graphics in entertainment industry, the numbers of digital films and animations is increased and the scenes of numerous crowd are shown to us. Though there are many studies on the techniques to process the behavior of crowd effectively and the developments of crowd behavioral systems, there is not enough study on the development for an efficient crowd behavioral simulator. In this' paper, we smartly present the types offish behavior in cyber underwater and make up for the weak points of time and cost. We develop the fish schooling behavior simulator for the contents of cyber underwater, automating fish behavioral types realistically and efficiently.

• Information and Communications Magazine
• /
• v.29 no.4
• /
• pp.62-71
• /
• 2012

수십 억년 동안 진화를 거듭해온 지구상의 생명체들은 외부의 제어 없이 독자적으로 단순한 행동 규칙에 따라 기능을 수행하여 주어진 목적의 최적해를 달성한다. 이러한 다양한 생명체의 행동 원리를 모델링하여 만든 알고리즘을 생체모방 알고리즘(Bio-Inspired Algorithm)이라 한다. 생체모방 알고리즘은 다수의 개체가 존재하며, 주변 환경이 동적으로 변하고, 가용 자원의 제약이 주어지며, 이질적인 특성을 갖는 개체들이 분잔 및 자율적으로 움직이는 환경에서 안정성, 확장성, 적응성과 같은 특징을 보여주는데, 이는 통신 네트워크 환경 및 서비스 요구사항과 유사성을 갖는다. 본 논문에서는 대표적인 생체모방 알고리즘으로 통신 및 네트워킹 기술로 사용되는 Ant Colony 알고리즘, Bee 알고리즘, Firefly 알고리즘, Flocking 알고리즘에 대해 살펴보고, 관련 프로젝트 및 연구 동향을 정리한다. 이를 통해 현재의 생체모방 알고리즘의 한계를 극복하고 미래 통신 및 네트워킹 기술이 나아갈 방향을 제시한다.

• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.117-129
• /
• 2024

This paper introduces collective navigation through a narrow gap using a curriculum-based deep reinforcement learning algorithm for a swarm of unmanned aerial vehicles (UAVs). Collective navigation in complex environments is essential for various applications such as search and rescue, environment monitoring and military tasks operations. Conventional methods, which are easily interpretable from an engineering perspective, divide the navigation tasks into mapping, planning, and control; however, they struggle with increased latency and unmodeled environmental factors. Recently, learning-based methods have addressed these problems by employing the end-to-end framework with neural networks. Nonetheless, most existing learning-based approaches face challenges in complex scenarios particularly for navigating through a narrow gap or when a leader or informed UAV is unavailable. Our approach uses the information of a certain number of nearest neighboring UAVs and incorporates a task-specific curriculum to reduce learning time and train a robust model. The effectiveness of the proposed algorithm is verified through an ablation study and quantitative metrics. Simulation results demonstrate that our approach outperforms existing methods.

• Journal of Korea Multimedia Society
• /
• v.6 no.2
• /
• pp.332-339
• /
• 2003

To solve the large scale optimization problem approximately, various approaches have been introduced. Recently the Particle Swarm Optimization has been introduced. The Particle Swarm Optimization simulates the process of birds flocking or fish schooling for food, as with the information of each agent is skated by other agents. The Particle Swarm Optimization technique has been applied to various optimization problems whose variables are continuous. However, there are seldom trials for visualization of searching process. This paper proposes a new visualization tool for searching process of Particle Swarm Optimization algorithm. The proposed tool is effective for understanding the searching process of Particle Swarm Optimization method and educational for students. The computational results can be shown tiny and very helpful for education.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.2 no.1
• /
• pp.25-33
• /
• 2007

Crowd behaviors on cyber underwater environment are often produced in entertainment contents, such as films and games. It is easy for us to come in contact with the scenes appearing a lot of characters as digital films and animation works are increased gradually, owing to developing of computer graphics. Though the processing a scene of crowd and the behavior system of crowd, related to the processing techniques of crowd behavior in cyber space, have been implemented so far, the research for developing the natural crowd behavior simulator can not be still satisfying. In this paper, we designed a realistic and efficient Fish Schooling Behavior Simulator for the contents production of cyber underwater environment, which showed each type of fish behavior in cyber underwater smartly, and which generated the animating the behavior automatically, reducing the time and cost.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1435-1440
• /
• 2021

With the advent of the low interest rate era, many investors are flocking to the stock market. In the past stock market, people invested in stocks labor-intensively through company analysis and their own investment techniques. However, in recent years, stock investment using artificial intelligence and data has been widely used. The success rate of stock prediction through artificial intelligence is currently not high, so various artificial intelligence models are trying to increase the stock prediction rate. In this study, we will look at various artificial intelligence models and examine the pros and cons and prediction rates between each model. This study investigated as stock prediction programs using artificial intelligence artificial neural network (ANN), deep learning or hierarchical learning (DNN), k-nearest neighbor algorithm(k-NN), convolutional neural network (CNN), recurrent neural network (RNN), and LSTMs.