• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.179-186
• /
• 2021

Manipulation of complex objects with an anthropomorphic robot hand like a human hand is a challenge in the human-centric environment. In order to train the anthropomorphic robot hand which has a high degree of freedom (DoF), human demonstration augmented deep reinforcement learning policy optimization methods have been proposed. In this work, we first demonstrate augmentation of human demonstration in deep reinforcement learning (DRL) is effective for object manipulation by comparing the performance of the augmentation-free Natural Policy Gradient (NPG) and Demonstration Augmented NPG (DA-NPG). Then three DRL policy optimization methods, namely NPG, Trust Region Policy Optimization (TRPO), and Proximal Policy Optimization (PPO), have been evaluated with DA (i.e., DA-NPG, DA-TRPO, and DA-PPO) and without DA by manipulating six objects such as apple, banana, bottle, light bulb, camera, and hammer. The results show that DA-NPG achieved the average success rate of 99.33% whereas NPG only achieved 60%. In addition, DA-NPG succeeded grasping all six objects while DA-TRPO and DA-PPO failed to grasp some objects and showed unstable performances.

• Journal of the Korea Society for Simulation
• /
• v.18 no.3
• /
• pp.23-34
• /
• 2009

In the railroad rolling stock depot, long-term maintenance tasks is done regularly every two or four year basis to maintain the functionality of equipments and rolling stock body or for the repair operation of the heavily damaged rolling stocks by fatal accidents. This paper addresses the computer simulation model building for the rolling stock maintenance shop for the CDC(Commuter Diesel Car) and Generator Car planned to be constructed at Daejon Rolling Stock Depot, which will be moved from Yongsan Rolling Stock Depot. We evaluated the processing capacity of two layout design alternatives based on the maintenance process chart through the developed simulation models. The performance measures are the number of processed cars per year, the cycle time, shop utilization, work in process and the average number waiting car for input. The simulation result shows that one design alternative outperforms another design alternative in every aspect and superior design alternative can process total 340 number of trains per year 15% more than the proposed target within the current average cycle time.

• Journal of KIISE:Databases
• /
• v.31 no.5
• /
• pp.527-539
• /
• 2004

As more and more documents are published in XML, generating relational schemas to store XML documents in a relational database is also getting important. This paper describes a technique as well as its implementation to produce a relational schema from the XML Schema, a standard recently recommended by W3C. The DTD-based inlining technique cannot be applied to the XML Schema, because the XML Schema has many new features, which don't exist in the DTD. Various built-in data types, inheritance, and polymorphism, for example, strengthen the XML Schema, but make the generation of a relational schema from an XML Schema more difficult. We propose an XML Schema Inlining Technique, based on the previous work. The technique first maps various data types in the XML Schema to those of the relational database. After that, it construct the schema graph and the type graph from types and elements defined in the XML Schema. The relational schema is generated while traversing the type graphs. Besides, we describe techniques for handling xsi：type, used for the polymorphism, and the anon#moos type. We also propose a couple of heuristic methods for enhancing the performance of the system. Finally, we conducted experiments to show that our technique is better than the binary table approach.

• The KIPS Transactions:PartB
• /
• v.17B no.5
• /
• pp.345-354
• /
• 2010

Visual tracking techniques for Augmented Reality are classified as either a marker tracking approach or a natural feature tracking approach. Marker-based tracking algorithms can be efficiently implemented sufficient to work in real-time on mobile devices. On the other hand, natural feature tracking methods require a lot of computationally expensive procedures. Most previous natural feature tracking methods include heavy feature extraction and pattern matching procedures for each of the input image frame. It is difficult to implement real-time augmented reality applications including the capability of natural feature tracking on low performance devices. The required computational time cost is also in proportion to the number of patterns to be matched. To speed up the natural feature tracking process, we propose a novel fast tracking method based on optical flow. We implemented the proposed method on mobile devices to run in real-time and be appropriately used with mobile augmented reality applications. Moreover, during tracking, we keep up the total number of feature points by inserting new feature points proportional to the number of vanished feature points. Experimental results showed that the proposed method reduces the computational cost and also stabilizes the camera pose estimation results.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.1
• /
• pp.59-69
• /
• 2018

In this paper, we present a robust beamforming design to tackle the weighted sum-rate maximization (WSRM) problem in a multicell multiple-input multiple-output (MIMO) - non-orthogonal multipleaccess (NOMA) downlink system for 5G wireless communications. This work consider the imperfectchannel state information (CSI) at the base station (BS) by adding uncertainties to channel estimation matrices as the worst-case model i.e., singular value uncertainty model (SVUM). With this observation, the WSRM problem is formulated subject to the transmit power constraints at the BS. The objective problem is known as on-deterministic polynomial (NP) problem which is difficult to solve. We propose an robust beam forming design which establishes on majorization minimization (MM) technique to find the optimal transmit beam forming matrix, as well as efficiently solve the objective problem. In addition, we also propose a joint user clustering and power allocation (JUCPA) algorithm in which the best user pair is selected as a cluster to attain a higher sum-rate. Extensive numerical results are provided to show that the proposed robust beamforming design together with the proposed JUCPA algorithm significantly increases the performance in term of sum-rate as compared with the existing NOMA schemes and the conventional orthogonal multiple access (OMA) scheme.

• (The)Study of the Eastern Classic
• /
• no.56
• /
• pp.241-265
• /
• 2014

This paper looked at the existing situation and the uniqueness of the Hasu(河西) kim Inhu(金麟厚)'s collection of works. As a result, The first edition includes a piece missing in the second edition. For the title of the same work, the second edition differs from the third edition. And in the third edition, The missing piece is found through the annual report. This research found the performance that were found missing pieces. But I did not review for that depth meaning, and postpone it to a later date.

• Journal of the Korean Society for information Management
• /
• v.39 no.4
• /
• pp.269-306
• /
• 2022

This study aimed to develop indicators that can measure the digital transformation performance of science and technology information construction and sharing systems by utilizing the Digital Curation Maturity Models. For digital transformation, it is necessary to consider not only simple service improvement but also organizational and business changes. In this study, we aimed to develop a model for measuring the digital transformation of KISTI, Korea's representative science and technology information service organization. KISTI has already carried out BPR work for digital transformation and borrowed the concept of a maturity model. However, in BPR, there is no method to measure the result. Therefore, in this paper, we developed an index to measure digital transformation based on the maturity model. Indicator development was carried out in two ways: model development and evaluation. Cases for model construction were made through a comprehensive review of existing KISTI and various domestic and foreign cases. The models before verification were technology (37), data (45), strategy (18), organization (36), and (social)influence (14) based on the major categories. After verification using confirmatory factor analysis, the model is classified as technology (20 / 17 indicators dropped), data (36 / 9 indicators dropped), strategy (18 / maintenance), organization(30 / 6 indicators dropped), and (social) influence (13 indicators / 1 indicator dropped).

• Fire Science and Engineering
• /
• v.20 no.2 s.62
• /
• pp.31-43
• /
• 2006

In Korea, the Emergency Medical Service, EMS is provided by Fire Services with 119 EMS and all the public uses the service for free. Although it appears very successful and is respected nationally, structural problem EMS exist and are worsening. First of all, the "free riding effect" becomes increasingly problematic. Some argue that 30% of the demands is not urgent or emergent. The total number of demands is increasing even without the free riding effect. The Current EMS system itself cannot meet the increasing EMS demand. The medical aspect is so poor that the EMS cannot dispatch a medical specialist to the scene. The cardiac arrest resuscitation rate is only $1.24{\sim}9.9%$, compared to 40% in Boston, MA, USA. But due to the regulations and limitations of the Fire Service organization, it is difficult to secure enough EMS resources. To work out these problems, it needs a structural innovation. To secure enough resources and achieve higher medical performance we should invite the medical sector and the private sector into EMS arena by contracting partnerships with Fire Services and charging a reasonable EMS fee. We found through statistical test that any partnership system is more effective than fire-alone system and most countries around the world have partnership system rather than fire-alone system.

• The KIPS Transactions:PartC
• /
• v.14C no.6
• /
• pp.519-524
• /
• 2007

One of the most important issues on the sensor network with resource limited sensor nodes is prolonging the network lifetime by effectively utilizing the limited node energy. The most representative mechanism to achieve a long lived sensor network is the clustering mechanism which can be further classified into the single hop mode and the multi hop mode. The single hop mode requires that all sensor nodes in a cluster communicate directly with the cluster head(CH) via single hop md, in the multi hop mode, sensor nodes communicate with the CH with the help of other Intermediate nodes. One of the most critical factors that impact on the performance of the existing multi hop clustering mechanism is the cluster size and, without the assumption on the uniform node distribution, finding out the best cluster size is intractable. Since sensor nodes in a real sensor network are distributed non uniformly, the fixed size mechanism may not work best for real sensor networks. Therefore, in this paper, we propose a new dynamic size multi hop clustering mechanism in which the cluster size is determined according to the distance from the sink to relieve the traffic passing through the CHs near the sink. We show that our proposed scheme outperforms the existing fixed size clustering mechanisms by carrying out numerical analysis and simulations.

• The Journal of the Korea Contents Association
• /
• v.14 no.1
• /
• pp.423-434
• /
• 2014

This study examined the relative effects of positive and negative reinforcement on customer service behaviors. In addition, we examined whether the intervention would lead to response generalization on non-target behaviors. Five employees at three different convenient stores participated and ABC/ACB within-subject design was adopted. For the three participants, followed by the baseline(A), the positive reinforcement was first introduced(B) and the negative reinforcement(C) was introduced for the next phase. For the remaining two participants, the negative reinforcement(C) was first introduced after baseline(A) and the positive reinforcement (B) was introduced. Results showed a greater improvement in target behaviors under the positive reinforcement condition than that of the negative reinforcement condition. In addition, both reinforcement condition cause response generalization on non-targeted service behaviors, however, the comparable effects was found between two reinforcement conditions. Post-interview indicated that participants experienced positive emotions under positive reinforcement condition and negative emotions under negative reinforcement condition. These results suggest that the he techniques using positive reinforcement can be more effective and efficient to improve the work performance.