• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.91-93
• /
• 2009

In this paper, we present the Q-learning method for adaptive traffic signal control on the basis of In this paper, we present the Q-learning method for adaptive traffic signal control on the basis of multi-agent technology. The structure is composed of sixphase agents and one intersection agent. Wireless communication network provides the possibility of the cooperation of agents. As one kind of reinforcement learning, Q-learning is adopted as the algorithm of the control mechanism, which can acquire optical control strategies from delayed reward; furthermore, we adopt dynamic learning method instead of static method, which is more practical. Simulation result indicates that it is more effective than traditional signal system.

• Spatial Information Research
• /
• v.22 no.2
• /
• pp.89-97
• /
• 2014

This study investigated how much the spread of traffic control information affect the traffic congestion in order to identify the behavior of the individual drivers that impacts on the entire transport system. For this purpose, agent-based transportation model was constructed. GIS data were directly used for the transportation model and the processing steps of the simulation results are presented. The results showed that the average speed was not lowered when the traffic information was provided to 30 to 70% of total drivers. In contrast, the driver's average speed is reduced when he traffic information was provided to less than 20% or 80% or more. In summary, the provision of traffic information to drivers has an influence on the traffic flow and bypassing vehicles can generate local congestion. This results can be used as a basis for the future direction of road transport policy.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.331-333
• /
• 2004

본 논문에서는 SIP(Session initiation Protocol) 기반의 VoIP(Voice over Internet Protocol) 시스템에 효율적인 call-control을 위해 필요한 헤더와 파라미터를 추가한 Extension SIP를 제안하였다. 또한 이 제시된 Extension SIP에 따르는 SIP Proxy Server와 User Agent(User Agent Client, User Agent Server)를 리눅스 시스템에서 C언어를 통해 구현하였고, 이 구현된 Extension SIP 시스템을 통해 기존의 SIP 시스템과 cail-control을 위한 packet traffic을 비교.분석 함으로써 제안한 Extension SIP의 효율성 을 증명하였다.

• Journal of Navigation and Port Research
• /
• v.44 no.6
• /
• pp.469-476
• /
• 2020

Modeling and Simulation (M&S) systems which deal with situational complexity often require human involvement due to the high-level decision-making that is necessary for ship movement, navigation, control center management, shipping company logistics, meteorological system information, and maritime transportation GIS. In order to properly simulate maritime traffic, it is necessary to accurately model the human decision-making process of the ship officer, including aspects of the ship officer's behavioral tendencies, personal navigation experience, and pattern of voyage errors, as this is the most accurate way in which to reproduce and predict realistic maritime traffic conditions. In this paper, which looks at agent-based maritime traffic simulation, we created a basic survey in order to conduct behavior analysis on ship operators' collision avoidance strategies. Using the information gathered throughout the survey, we developed an agent-based navigational behavior model which attempts to capture the behavioral patterns of a ship officer during an instance of ship collision. These results could be used in the future in further developments for more advanced maritime traffic simulation.

• Nuclear Engineering and Technology
• /
• v.53 no.7
• /
• pp.2195-2206
• /
• 2021

In order to mitigate the damage caused by accidents in nuclear power plants (NPPs), evacuation strategies are usually managed on the basis of off-site effects such as the diffusion of radioactive materials and evacuee traffic simulations. However, the interactive behavior between evacuees and the accident environment has a significant effect on the consequential gap. Agent-based modeling (ABM) is a method that can control and observe such interactions by establishing agents (i.e., the evacuees) and patches (i.e., the accident environments). In this paper, a radiological emergency evacuation model is constructed to realistically check the effectiveness of an evacuation strategy using NetLogo, an ABM toolbox. Geographic layers such as radiation sources, roads, buildings, and shelters were downloaded from an official geographic information system (GIS) of Korea, and were modified into respective patches. The dispersion model adopted from the puff equation was also modified to fit the patches on the geographic layer. The evacuees were defined as vehicle agents and a traffic model was implemented by combining the shortest path search (determined by an A ^* algorithm) and a traffic flow model incorporated in the Nagel-Schreckenberg cellular automata model. To evaluate the radiological harm to the evacuees due to the spread of radioactive materials, a simple exposure model was established to calculate the overlap fraction between the agents and the dispersion patches. This paper aims to demonstrate that the potential of ABM can handle disaster evacuation strategies more realistically than previous approaches.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.21-25
• /
• 2004

The proposed QoS Agent System for distance video learning is used unicast and P2P. This method is developed for decreased the traffic of server by control agent. First step capture from stream server's camera, and transmitted streaming data to the first CSplayer(c/s), and outrank CSplayer transmitted streaming data to the subordinate CSplayer controled by QoS Control Agent.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.118-120
• /
• 2009

In this paper, we present the Q-learning method for adaptive traffic signal control on the basis of multi-agent technology. The structure is composed of sixphase agents and one intersection agent. Wireless communication network provides the possibility of the cooperation of agents. As one kind of reinforcement learning, Q-learning is adopted as the algorithm of the control mechanism, which can acquire optical control strategies from delayed reward; furthermore, we adopt dynamic learning method instead of static method, which is more practical. Simulation result indicates that it is more effective than traditional signal system.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.194-207
• /
• 2022

In the last decade, substantial progress has been achieved in intelligent traffic control technologies to overcome consistent difficulties of traffic congestion and its adverse effect on smart cities. Edge computing is one such advanced progress facilitating real-time data transmission among vehicles and roadside units to mitigate congestion. An edge computing-based deep reinforcement learning system is demonstrated in this study that appropriately designs a multiobjective reward function for optimizing different objectives. The system seeks to overcome the challenge of evaluating actions with a simple numerical reward. The selection of reward functions has a significant impact on agents' ability to acquire the ideal behavior for managing multiple traffic signals in a large-scale road network. To ascertain effective reward functions, the agent is trained withusing the proximal policy optimization method in several deep neural network models, including the state-of-the-art transformer network. The system is verified using both hypothetical scenarios and real-world traffic maps. The comprehensive simulation outcomes demonstrate the potency of the suggested reward functions.

• Journal of Korean Society of Transportation
• /
• v.24 no.5 s.91
• /
• pp.19-33
• /
• 2006

This paper presents the results of an experimental study on the application of NTCIP protocol to Korean traffic signal control system. For this study the communication Protocol of the existing traffic signal control system was adjusted to meet NTCIP standard. Management information base for Korea real-time traffic signal control system, message library of OER, traffic control center management software supporting SNMP/SFMP Protocol, and agent softwares for local controllers were developed during the experimental study. The applicability test of the adjusted system by NTCIP standard was performed. Fifty eight Percent of communication packets were lost at 2.400bps communication speed, which made the operation impossible. The experimentations with communication speeds 4,800bps and 9,600bps did not cause problems. In conclusion, to apply the NTCIP standard to domestic real-time traffic control system, communication environments need to be upgraded to 4,800bps or higher.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.16 no.2
• /
• pp.44-49
• /
• 2007

An agent is an abstract unit for problem solving in the field of distributed artificial intelligence, and an agent-based system is designed and implemented based on the definition of agent as its central concept. Agent modeling is advantageous to abstraction, disintegration and structuring for describing complex system, so its application is increased in various areas including air traffic control, power transmission, e-commerce and medicine. There is no agreed definition of agent but agents have common points as follows: autonomy, reactivity, pro-activeness and cooperation. An agent-oriented modeling is an approach of a concept different form existing object-oriented modeling. This study proposed the agent application for E-Beam manufacturing system. To evaluate the performance of the proposed process design, we used the JADE library. The JADE toolkit provides a FIPA-compliant agent platform and a package to develp Java agents. It provides a basic set of functionalities that are regarded as essential for an autonomous agent architecture.