• International Journal of Control, Automation, and Systems
• /
• v.6 no.2
• /
• pp.253-262
• /
• 2008

This paper presents a framework for decentralized control of self-organizing swarm systems based on the artificial potential functions (APFs). The framework explores the benefits by associating agents based on position information to realize complex swarming behaviors. A key development is the introduction of a set of association rules by APFs that effectively deal with a host of swarming issues such as flexible and agile formation. In this scheme, multiple agents in a swarm self-organize to flock and achieve formation control through attractive and repulsive forces among themselves using APFs. In particular, this paper presents an association rule for swarming that requires less movement for each agent and compact formation among agents. Extensive simulations are presented to illustrate the viability of the proposed framework.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.236-240
• /
• 2000

A cellular automaton is well-known for self-organizing and dynamic behaviors in the field of artificial life. This paper addresses a new neuronic architecture called an evolvable cellular classifier which evolves with the genetic rules (chromosomes) in the non-uniform cellular automata. An evolvable cellular classifier is primarily based on cellular programing, but its mechanism is simpler because it utilizes only mutations for the main genetic operators and resembles the Hopfield network. Therefore, the desirable hi t-patterns could be obtained through evolutionary processes for just one individual agent. As a result, an evolvable hardware is derived which is applicable to classification of bit-string information.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.01a
• /
• pp.243-246
• /
• 2018

본 논문에서는 2차원 공간상에서 상부에서 하부로 떨어지는 블록을 하부에서 피하는 게임에서 강화 학습에 사용되는 DQN 알고리즘을 이용하여 블록 피하기 에이전트를 구현하고 학습 통해 점점 더 높은 점수를 받는 모습을 확인하였다. 파이썬을 이용하여 게임을 구현한 후 텐서플로우를 이용하여 DQN를 이용한 에이전트를 구현하였다. 에이전트는 보상을 통한 학습을 통하여 점점 강화되도록 하였는데, 초기에는 무작위로 움직였으나, 환경으로부터 받는 보상으로 점점 더 능숙하게 피하는 모습을 관찰할 수 있었다. 본 구현에서는 4000번 정도의 게임 시행에서 아주 능숙하게 피하는 결과를 얻을 수 있었다.

• The Korean Journal of Mycology
• /
• v.49 no.3
• /
• pp.413-416
• /
• 2021

In July 2019, damping-off symptoms of young Fischer's ragwort (Ligularia fischeri) plants were observed in four vinyl greenhouses of a farmer located in Taebaek, Gangwon Province, Korea. This disease occurred in 20-30% of plants in the vinyl greenhouses investigated. Nine isolates of Rhizoctonia sp. were obtained from the diseased plants. All the isolates were identified as Rhizoctonia solani AG-2-2 (IIIB), based on morphological, cultural characteristics, and anastomosis test. Three isolates were used for artificial inoculation test on Fischer's ragwort. Pathogenicity of these isolates was confirmed on the plants with the inoculation tests. Damping-off symptoms observed on the inoculated plants were similar to those observed in the diseased plants in the vinyl greenhouses. This is the first report of R. solani AG-2-2 (IIIB) being the causative agent in damping-off in Fischer's ragwort.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.132-140
• /
• 2021

Under the mobile ad-hoc network system, the main reason for causing congestion is because of the limited availability of resources. On the other hand, the standardised TCP based congestion controlling mechanism is unable to control and handle the major properties associated with the shared system of wireless channels. It creates an effect on the design associated with suitable protocols along with protocol stacks through the process of determining the mechanisms of congestion on a complete basis. Moreover, when bringing a comparison with standard TCP systems the major environment associated with mobile ad hoc network is regraded to be more problematic on a complete basis. On the other hand, an agent-based mobile technique for congestion is designed and developed for the part of avoiding any mode of congestion under the ad-hoc network systems.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.3
• /
• pp.110-113
• /
• 2022

Non-playable game character plays an important role in improving the concentration of the game and the interest of the user, and recently implementation of NPC with reinforcement learning has been in the spotlight. In this paper, we estimate an AI target tracking method via reinforcement learning, and implement an AI-based tracking agency of specific target object with avoiding traps through Unity ML-Agents Toolkit. The implementation is built in Unity game engine, and simulations are conducted through a number of experiments. The experimental results show that outstanding performance of the tracking target with avoiding traps is shown with good enough results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.47-49
• /
• 2022

Robot navigation has seen a major improvement since the the rediscovery of the potential of Artificial Intelligence (AI) and the attention it has garnered in research circles. A notable achievement in the area was Deep Learning (DL) application in computer vision with outstanding daily life applications such as face-recognition, object detection, and more. However, robotics in general still depend on human inputs in certain areas such as localization, navigation, etc. In this paper, we propose a study case of robot navigation based on deep reinforcement technology. We look into the benefits of switching from traditional ROS-based navigation algorithms towards machine learning approaches and methods. We describe the state-of-the-art technology by introducing the concepts of Reinforcement Learning (RL), Deep Learning (DL) and DRL before before focusing on visual navigation based on DRL. The case study preludes further real life deployment in which mobile navigational agent learns to navigate unbeknownst areas.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.442-445
• /
• 2023

최근 딥러닝 및 로봇기술의 발전으로 인해 대량의 데이터를 빠르게 수집하고 처리하는 연구 분야들로 확대되었다. 이와 관련된 한 가지 분야로써 다중 로봇을 이용한 분산학습 연구가 있으며, 이는 단일 에이전트를 이용할 때보다 대량의 데이터를 빠르게 수집 및 처리하는데 용이하다. 본 연구에서는 기존 Distributed Neural Network Optimization (DiNNO) 알고리즘에서 제안한 정적 분산 학습방법과 달리 단계적 분산학습 방법을 새롭게 제안하였으며, 모델 성능을 향상시키기 위해 원시 변수를 근사하는 단계수를 상수로 고정하는 기존의 방식에서 통신회차가 늘어남에 따라 점진적으로 근사 횟수를 높이는 방법을 고안하여 새로운 알고리즘을 제안하였다. 기존 알고리즘과 제안된 알고리즘의 정성 및 정량적 성능 평가를 수행하기 MNIST 분류와 2 차원 평면도 지도화 실험을 수행하였으며, 그 결과 제안된 알고리즘이 기존 DiNNO 알고리즘보다 동일한 통신회차에서 높은 정확도를 보임과 함께 전역 최적점으로 빠르게 수렴하는 것을 입증하였다.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1065-1078
• /
• 2023

In vehicular networks, diverse safety information can be shared among vehicles through internet connections. In vehicle-to-internet communications, vehicles on the road are wirelessly connected to different cloud networks, thereby accelerating safety information exchange. Onboard sensors acquire traffic-related information, and reliable intermediate nodes and network services, such as navigational facilities, allow to transmit safety information to distant target vehicles and stations. Using vehicle-to-network communications, we minimize delays and achieve high accuracy through consistent connectivity links. Our proposed approach uses intermediate nodes with two-hop separation to forward information. Target vehicle detection and routing of safety information are performed using machine learning algorithms. Compared with existing vehicle-to-internet solutions, our approach provides substantial improvements by reducing latency, packet drop, and overhead.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.5
• /
• pp.1609-1623
• /
• 2014

A number of studies have been conducted recently regarding the development of automation systems for the construction sector. Much of this attention has focused on earthwork because it is highly dependent on construction machines and is regarded as being basic for the construction of buildings and civil works. For example, technologies are being developed in order to enable earthwork planning based on construction site models that are constructed by automatic systems and to enable construction equipment to perform the work based on the plan and the environment. There are many problems that need to be solved in order to enable the use of automatic earthwork systems in construction sites. For example, technologies are needed for enabling collaborations between similar and different kinds of construction equipment. This study aims to develop a construction system that imitates collaborative systems and decision-making methods that are used by humans. The proposed system relies on the multi-agent concept from the field of artificial intelligence. In order to develop a multi-agent-based system, configurations and functions are proposed for the agents and a framework for collaboration and arbitration between agents is presented. Furthermore, methods are introduced for preventing duplicate work and minimizing interference effects during the collaboration process. Methods are also presented for performing advance planning for the excavators and compactors that are involved in the construction. The current study suggests a theoretical framework and evaluates the results using virtual simulations. However, in the future, an empirical study will be conducted in order to apply these concepts to actual construction sites through the development of a physical system.