• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.245-248
• /
• 2000

In this paper we improve the performance of autonomous mobile robot by induction of reinforcement learning concept. Generally, the system used in this paper is divided into two part. Namely, one is neural-fuzzy and the other is dynamic recurrent neural networks. Neural-fuzzy determines the next action of robot. Also, the neural-fuzzy is determined to optimal action internal reinforcement from dynamic recurrent neural network. Dynamic recurrent neural network evaluated to determine action of neural-fuzzy by external reinforcement signal from environment, Besides, dynamic recurrent neural network weight determined to internal reinforcement signal value is evolved by genetic algorithms. The architecture of propose system is applied to the computer simulations on controlling autonomous mobile robot.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.5
• /
• pp.558-568
• /
• 2021

The existing underwater vehicle controller design is applied by linearizing the nonlinear dynamics model to a specific motion section. Since the linear controller has unstable control performance in a transient state, various studies have been conducted to overcome this problem. Recently, there have been studies to improve the control performance in the transient state by using reinforcement learning. Reinforcement learning can be largely divided into value-based reinforcement learning and policy-based reinforcement learning. In this paper, we propose the roll controller of underwater vehicle based on Deep Deterministic Policy Gradient(DDPG) that learns the control policy and can show stable control performance in various situations and environments. The performance of the proposed DDPG based roll controller was verified through simulation and compared with the existing PID and DQN with Normalized Advantage Functions based roll controllers.

• Journal of Korean Association for Spatial Structures
• /
• v.21 no.2
• /
• pp.41-48
• /
• 2021

A smart tuned mass damper (TMD) is widely studied for seismic response reduction of various structures. Control algorithm is the most important factor for control performance of a smart TMD. This study used a Deep Deterministic Policy Gradient (DDPG) among reinforcement learning techniques to develop a control algorithm for a smart TMD. A magnetorheological (MR) damper was used to make the smart TMD. A single mass model with the smart TMD was employed to make a reinforcement learning environment. Time history analysis simulations of the example structure subject to artificial seismic load were performed in the reinforcement learning process. Critic of policy network and actor of value network for DDPG agent were constructed. The action of DDPG agent was selected as the command voltage sent to the MR damper. Reward for the DDPG action was calculated by using displacement and velocity responses of the main mass. Groundhook control algorithm was used as a comparative control algorithm. After 10,000 episode training of the DDPG agent model with proper hyper-parameters, the semi-active control algorithm for control of seismic responses of the example structure with the smart TMD was developed. The simulation results presented that the developed DDPG model can provide effective control algorithms for smart TMD for reduction of seismic responses.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.63-68
• /
• 2022

Autonomous driving technology has recently made great progress with the introduction of deep reinforcement learning. In order to effectively use deep reinforcement learning, it is important to select the appropriate activation function. In the meantime, many activation functions have been presented, but they show different performance depending on the environment to be applied. This paper compares and evaluates the performance of 12 activation functions to see which activation functions are effective when using reinforcement learning to learn autonomous driving on highways. To this end, a performance evaluation method was presented and the average reward value of each activation function was compared. As a result, when using GELU, the highest average reward could be obtained, and SiLU showed the lowest performance. The average reward difference between the two activation functions was 20%.

• Journal of the Korean Society of Safety
• /
• v.38 no.2
• /
• pp.44-51
• /
• 2023

With the recent increase in gas energy use, risk management for explosion accidents has been emphasized. Protective walls can be used to reduce damage from explosions. The KOSHA GUIDE D-65-2018 suggests the minimum thickness and height of protective walls, minimum reinforcement diameter, and maximum spacing of reinforcements for the structural safety of the protective walls. However, no related evidence has been presented. In this study, the blast load carrying capacity of the protective wall was analyzed by the pressure-impulse diagrams while changing the yield strength of the reinforcement, concrete compressive strength, reinforcement ratio, protective wall height, and thickness, to check the adequacy of the KOSHA GUIDE. Results show that failure may occur even with design based on the criteria presented by KOSHA GUIDE. In order to achieve structural safety of protective walls, additional criteria for minimum reinforcement yield strength and maximum height of protective wall are suggested for inclusion in KOSHA GUIDE. Moreover, the existing value for minimum reinforcement ratio and the thickness of the protective wall should be increased.

• Journal of the Korean Institute of Rural Architecture
• /
• v.24 no.2
• /
• pp.45-52
• /
• 2022

Of the 17,106 domestic reservoirs(as of December 2020), 14,611 are older than 50 years, and these old reservoirs will gradually increase over time. The injection grouting method is most applied to the reinforcement method of the aging reservoir. However, the injection grouting method is not accurate in uniformity and reinforced area. An laboratory model test was conducted to evaluate the applicability of the deep mixing method, which compensated for these shortcomings, as a reservoir reinforcement method. As a result of calculating the hydraulic conductiveity for each method through the model test results, the injection grouting method was calculated as a hydraulic conductiveity value that was about 7.5 times larger than that of the deep mixing method. As a result of measuring the water level change in the laboratory model test, it was found that the water level change decreased in the injection method and deep mixing method compared to the non-reinforcement method. In addition, deep mixing method showed a water level change of about 15% based on 40 hours compared to the injection method, indicating that the water-reducing effect was superior to that of the injection method.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.6
• /
• pp.904-914
• /
• 2008

An action-selection-mechanism(ASM) has been proposed to work as a fully connected finite state machine to deal with sequential behaviors as well as to allow a state in the task program to migrate to any state in the task, in which a primitive node in association with a state and its transitional conditions can be easily inserted/deleted. Also, such a primitive node can be learned by a shortest path-finding-based reinforcement learning technique. Specifically, we define a behavioral motivation as having state-dependent value as a primitive node for action selection, and then sequentially construct a network of behavioral motivations in such a way that the value of a parent node is allowed to flow into a child node by a releasing mechanism. A vertical path in a network represents a behavioral sequence. Here, such a tree for our proposed ASM can be newly generated and/or updated whenever a new behavior sequence is learned. To show the validity of our proposed ASM, experimental results of a mobile robot performing the task of pushing- a- box-in to- a-goal(PBIG) will be illustrated.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.40 no.12
• /
• pp.71-82
• /
• 2003

In this paper, we present a new method to calculate reinforcement value in QoS routing algorithm targeted for real-time applications based on Ant algorithm to efficiently and effectively reinforce ant-like mobile agents to find the best route toward destination in a network regarding necessary QoS metrics. Simulation results show that the proposed method realizes QoS routing more efficiently and more adaptively than those of the existing method thereby providing better solutions for the best route selection for real-time application that has high priority on delay jitter and bandwidth.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1998.10a
• /
• pp.501-506
• /
• 1998

This paper is an experimental study on the flexural strength and ductility capacity of reinforced high performance concrete beams with the concrete which has compressive strength of 600~700kg/$\textrm{cm}^2$, slump value of 20~25cm and slump-flow value of 60~70cm. Total 8 beams with different tensile reinforcement ratio and pattern of loading were tested. Form the results of reinforced high performance concrete beams, the equivalent stress block parameters proposed by MacGregor et al. or New Zealand code are recommended to use. Also, an extreme fiber concrete compressive strain of reinforced high performance concrete beams are distributed 0.0033~0.0048. In reinforced high performance concrete beams, reinforcement ratio in order to insure curvature ductility index 2 and 4 propose by ACI code should be less than those of reinforced normal strength concrete beams.

• KIEAE Journal
• /
• v.2 no.1
• /
• pp.41-48
• /
• 2002

This study aims to investigate and evaluate the room acoustic designs of a multi purpose gymnasiums which do not use adjustable treatments in order to change the acoustical characteristics. Considering the main uses of gymnasium and auditorium, experiments were carried out using both nondirectional speakers on the stage and loudspeaker installed on the ceiling. The result from the study are as follows; Measured RT under unoccupied condition was a little longer than the expected value, therefore, in the case of occupied condition RT would be close to the optimum value. However, parameters that evaluate intelligibility and speech transmission property appeared to be low and have large differences depending on the measuring points, therefore, more effective sound reflecting surfaces and sound reinforcement systems should be considered.