• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.15-20
• /
• 2023

Machine learning is comprised of supervised learning, unsupervised learning and reinforcement learning as the type of data and processing mechanism. In this paper, as input and output are unclear and it is difficult to apply the concrete modeling mathematically, reinforcement learning method are applied for crawling robot in this paper. Especially, Q-Learning is the most effective learning technique in model free reinforcement learning. This paper presents a method to implement a crawling robot that is operated by finding the most optimal crawling method through trial and error in a dynamic environment using a Q-learning algorithm. The goal is to perform reinforcement learning to find the optimal two motor angle for the best performance, and finally to maintain the most mature and stable motion about EV3 Crawling robot. In this paper, for the production of the crawling robot, it was produced using Lego Mindstorms with two motors, an ultrasonic sensor, a brick and switches, and EV3 Classroom SW are used for this implementation. By repeating 3 times learning, total 60 data are acquired, and two motor angles vs. crawling distance graph are plotted for the more understanding. Applying the Q-learning reinforcement learning algorithm, it was confirmed that the crawling robot found the optimal motor angle and operated with trained learning, and learn to know the direction for the future research.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2002.10a
• /
• pp.593-598
• /
• 2002

Routinely, strength method for the determination of the nominal moment of reinforced concrete beam is assumed to also be suitable for strengthening beams with carbon fiber sheets since typically strengthening beams compromise 98% by volume of reinforced concrete. Flexural capacity of strengthening beam is absolutely dependent upon the type of reinforcement materials, amount of reinforcement, anchoring system, adhesion capacity between reinforcement material and concrete. Therefore, it might be incorrect to use strength method for analysis and design of strengthening beam without considering the differences in the load-deflection curves, mechanism of failure, state of stress distribution, failure strain of the reinforcement. An flexural analysis based on force equilibrium and strain comparability has been developed for strengthening beam. Systematic experimental investigations are compared with analytical results. Then, the adaptation of strength method for strengthening beam have also been discussed.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.66-69
• /
• 2003

This paper considers a reinforcement learning(RL) which deals with real environments. Most reinforcement learning studies have been made by simulations because real-environment learning requires large computational cost and much time. Furthermore, it is more difficult to acquire many rewards efficiently in real environments than in virtual ones. The most important requirement to make real-environment learning successful is the appropriate construction of the state space. In this paper, to begin with, I show the basic overview of the reinforcement learning under real environments. Next, 1 introduce a state-space construction method under real environmental which is State Partition Method. Finally I apply this method to a robot navigation problem and compare it with conventional methods.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2001.03a
• /
• pp.121-128
• /
• 2001

This is a case study of stability analysis and reinforcement design for the tunnel where the collapse of the entrance slops occured along the fault zone developed in the bed rock. According to the site investigation, the main factor of sliding is the influence of fault gouge and heavy rainfall. Considering the in-situ condition, the versatile reinforcement methods is needed, and so the close investigation on the site area was accompanied with the stability analysis of tunnel and slops. The FRP(Fiberglass reinforced plastic) grouting method improved the defect of Steel Umbrella Arch Method, such as oxidation, low work efficiency, the material's heavy weight, is adapted as the reinforcement methods.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.09a
• /
• pp.337-349
• /
• 2009

In recent years, the cutting and banking areas along the railway in Korea are exposed to the erosion problem during every year. The reinforcement is a composite construction material in which the strength of engineering fill is enhanced by the addition of strong tensile reinforcement in many different types. Various problems of the railway infrastructure have occurred due to the differential settlement, frost heaving, mud pumping, lack of bearing capacity, partially loss of embankment. In advanced countries, railway roadbed reinforcement is applied to solve these problems on railway roadbed. This paper presents the solution of such problems by means of the engineering works incorporated with railway reinforcement infrastructures such as geotextile bag method, existing grouting method, geocell, reinforced earth, soil nailing and so on.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.23-31
• /
• 2018

Recently, there have been many studies on machine learning. Among them, studies on reinforcement learning are actively worked. In this study, we propose a controller to control bicycle using DDPG (Deep Deterministic Policy Gradient) algorithm which is the latest deep reinforcement learning method. In this paper, we redefine the compensation function of bicycle dynamics and neural network to learn agents. When using the proposed method for data learning and control, it is possible to perform the function of not allowing the bicycle to fall over and reach the further given destination unlike the existing method. For the performance evaluation, we have experimented that the proposed algorithm works in various environments such as fixed speed, random, target point, and not determined. Finally, as a result, it is confirmed that the proposed algorithm shows better performance than the conventional neural network algorithms NAF and PPO.

• The KIPS Transactions:PartB
• /
• v.11B no.1
• /
• pp.77-82
• /
• 2004

Reinforcement learning takes reward about selecting action when agent chooses some action and did state transition in Present state. this can be the important subject in reinforcement learning as temporal-credit assignment problems. In this paper, by new meta heuristic method to solve hard combinational optimization problem, examine Ant-Q learning method that is proposed to solve Traveling Salesman Problem (TSP) to approach that is based for population that use positive feedback as well as greedy search. And, suggest Ant-TD reinforcement learning method that apply state transition through diversification strategy to this method and TD-error. We can show through experiments that the reinforcement learning method proposed in this Paper can find out an optimal solution faster than other reinforcement learning method like ACS and Ant-Q learning.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2003.06a
• /
• pp.83-91
• /
• 2003

Always slope according to construct road exist danger because of environment unstability factor in slope, Since this research an inclination of slope is gentle slope (1：1.5∼1：9.0) but falling happened by conduct of continuous ground movement. And this study considered more economical and efficient reinforcement method for slope stability. The various reinforcement methods are applied to execute examination of slope stability. Applied reinforcement methods satisfied safety factor And this research region is performing continuous measurement about ground movements and displacements.

• Structural Engineering and Mechanics
• /
• v.3 no.6
• /
• pp.593-609
• /
• 1995

The presence of reinforcement has a significant influence on the stress-strain behaviour of reinforced concrete structures, expecially when the failure stage of the structures is approached. In the present paper, the constrained and non-constrained zones of concrete due to the presence of reinforcement is developed and the stress-stress-strain behaviour of concrete is enhanced by a reinforcement confinement coefficient, Furthermore, a flexible method for the modelling of reinforcement with arbitrary orientation and not passing the nodes of concrete element is also proposed. Numerical examples and laboratory tests have shown that the coefficient and the modelling technique proposed by the author are satisfactory.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.18 no.1
• /
• pp.48-58
• /
• 2014

In this study, quasi-static according to the displacement-controlled (strain control) method tests on RC columns for seismic reinforcement performance in accordance with the provisions of the seismic design and construction before 1992 design code for highway bridges in korea. Used reinforcement that improves the performance of Inorganic Helical Bar, a kind of alloy steel, circular columns were tested outside the seismic reinforcing. In the experiment, fracture behavior, lateral load-displacement relation, ductility and energy assessment evaluation was performed through tests. The variables in experimental are section force of reinforcement, spiral reinforcement spacing, reinforcement method. Improved seismic performance and effect were confirmed through quasi-static test experiments. The results of study confirmed determination the appropriate size of reinforcement, reinforcement forces, spacing and selection of the type required, furthermore, not only mechanical reinforcement but also substitution of high-strength concrete reinforced with concrete cover improved seismic performance.