• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.2
• /
• pp.93-104
• /
• 2019

Minimal art, which began to flourish in the mid-1960s, explores perceptual situations caused by the involvement of objects in given site contexts. This has led to the mentions of minimal art as a site-specific art, but its limitations have also been pointed out. This study specifically addresses the limitations of minimal art as a site-specific art with two perceptual points of view. First, according to Michael Fried, situations described as 'now here' focus largely on the bodily experiences of a place. However, they do not rooted in specific time and space of a certain place. Second, the unique characteristics of a certain place are excluded from the perception of the body which occupies the passage of time. Self-sufficient algorithm, which is far from site-specific conditions, is the autonomous system creating the period in the way of arrangement of objects. In addition, Minimal art regards a body only as the objectivity excluding the subjectivity which is essential creating meaning in a place. In the latter part of the article, these features are dealt with through Donald Judd's works. This study on site-specificity also provides a new perspective on the discussion of Minimal architecture and Minimal landscape.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.8
• /
• pp.2993-3010
• /
• 2021

The job shop scheduling problem (JSSP) plays a critical role in smart manufacturing, an effective JSSP scheduler could save time cost and increase productivity. Conventional methods are very time-consumption and cannot deal with complicated JSSP instances as it uses one optimal algorithm to solve JSSP. This paper proposes an effective scheduler based on deep learning technology named self-supervised long-short term memory (SS-LSTM) to handle complex JSSP accurately. First, using the optimal method to generate sufficient training samples in small-scale JSSP. SS-LSTM is then applied to extract rich feature representations from generated training samples and decide the next action. In the proposed SS-LSTM, two channels are employed to reflect the full production statues. Specifically, the detailed-level channel records 18 detailed product information while the system-level channel reflects the type of whole system states identified by the k-means algorithm. Moreover, adopting a self-supervised mechanism with LSTM autoencoder to keep high feature extraction capacity simultaneously ensuring the reliable feature representative ability. The authors implemented, trained, and compared the proposed method with the other leading learning-based methods on some complicated JSSP instances. The experimental results have confirmed the effectiveness and priority of the proposed method for solving complex JSSP instances in terms of make-span.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.69-72
• /
• 1996

This paper considers a global resolution of kinematic redundancy under inequality constraints as a constrained optimal control. In this formulation, joint limits and obstacles are regarded as state variable inequality constraints, and joint velocity limits as control variable inequality constraints. Necessary and sufficient conditions are derived by using Pontryagin's minimum principle and penalty function method. These conditions leads to a two-point boundary-value problem (TPBVP) with natural, periodic and inequality boundary conditions. In order to solve the TPBVP and to find a global minimum, a numerical algorithm, named two-stage algorithm, is presented. Given initial joint pose, the first stage finds the optimal joint trajectory and its corresponding minimum performance cost. The second stage searches for the optimal initial joint pose with globally minimum cost in the self-motion manifold. The effectiveness of the proposed algorithm is demonstrated through a simulation with a 3-dof planar redundant manipulator.

• Smart Structures and Systems
• /
• v.29 no.6
• /
• pp.757-766
• /
• 2022

To train deep learning algorithms, a sufficient number of data are required. However, in most engineering systems, the acquisition of fault data is difficult or sometimes not feasible, while normal data are secured. The dearth of data is one of the major challenges to developing deep learning models, and fault diagnosis in particular cannot be made in the absence of fault data. With this context, this paper proposes an anomaly detection methodology for rotating machines using only normal data with self-labeling. Since only normal data are used for anomaly detection, a self-labeling method is used to generate a new labeled dataset. The overall procedure includes the following three steps: (1) transformation of normal data to self-labeled data based on a pretext task, (2) training the convolutional neural networks (CNN), and (3) anomaly detection using defined anomaly score based on the softmax output of the trained CNN. The softmax value of the abnormal sample shows different behavior from the normal softmax values. To verify the proposed method, four case studies were conducted, on the Case Western Reserve University (CWRU) bearing dataset, IEEE PHM 2012 data challenge dataset, PHMAP 2021 data challenge dataset, and laboratory bearing testbed; and the results were compared to those of existing machine learning and deep learning methods. The results showed that the proposed algorithm could detect faults in the bearing testbed and compressor with over 99.7% accuracy. In particular, it was possible to detect not only bearing faults but also structural faults such as unbalance and belt looseness with very high accuracy. Compared with the existing GAN, the autoencoder-based anomaly detection algorithm, the proposed method showed high anomaly detection performance.

• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.498-503
• /
• 1994

The paper demonstrates that holon networks can be used effectively for identification of nonlinear dynamical systems. The emphasis of the paper is on modeling of complicated systems which have a great deal of uncertainty and unknown interactions between their elements and parameters. The concept of applying a quantitative model building, for example, to environmental or ecological systems is not new. In a previous paper we presented a holon network model as an another alternative to quantitative modeling. Holon networks have a hierarchical construction where each level of hierarchy consists of networks with reciprocal actions among their elements. The networks are able to evolve by self-organizing their structure and adapt their parameters to environments. This was achieved by an autonomous decentralized adaptation algorithm. In this paper we propose a new emergent evolution algorithm. In this algorithm the initial holon networks consists of only a few elements and it grows gradually with each new observation in order to fit their function to the environment. Some examples show that this algorithm can lead to a network structure which has sufficient flexibility and adapts well to the environment.

• KSBB Journal
• /
• v.15 no.1
• /
• pp.14-21
• /
• 2000

In this study, algae cultivation using the photobioreactor has been applied to remove the nitrogen and phosphorus compounds in the wastewater of the livestock industry. The optimal ratio of nitrate and ortho-phosphate concentration was found for the enhancement of removal efficiency. To achieve the high density culture of algae, the photobioreactor consisted of optical fibers wes developed to get the sufficient light intensity. The light could be illuminated uniformly from light source to the entire reactor by the optical fibers. The structured kinetic model was proposed to describe the growth rate, consumption rate of nitrates and ortho-phosphates in algae culture. The self-organizing fuzzy logic controller incorporated with genetic algorithm was constructed to control the semi-continuous wastewater treatment system. The proposed fuzzy logic controller was applied to maintain the nitrated concentration at the given set-point with the control of wastewater feeding rate. The experimental results showed that the self-organizing fuzzy logic controller could keep the nitrate concentration and enhance algae growth.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.340-367
• /
• 2018

Storage services integration can be done for achieving high availability, improving data access performance and scalability while preventing vendor lock-in. However, multiple services environment management and interoperability have become a critical issue as a result of service architectures and communication interfaces heterogeneity. Storage federation model provides the integration of multiple heterogeneous and self-sufficient storage systems with a single control point and automated decision making about data distribution. In order to integrate diverse heterogeneous storage services into a single storage pool, we are proposing a storage service federation framework named BoxBroker. Moreover, an automated decision model based on a policy-driven data distribution algorithm and a service evaluation method is proposed enabling BoxBroker to make optimal decisions. Finally, a demonstration of our proposal capabilities is presented and discussed.

• Kyungpook Mathematical Journal
• /
• v.49 no.4
• /
• pp.701-712
• /
• 2009

In this paper, we first introduce a family S = {$S_n$ : C ${\rightarrow}$ C} of non-Lipschitzian mappings, called total asymptotically nonexpansive (briefly, TAN) on a nonempty closed convex subset C of a real Banach space X, and next give necessary and sufficient conditions for strong convergence of the sequence {$x_n$} defined recursively by the algorithm $x_{n+1}$ = $S_nx_n$, $n{\geq}1$, starting from an initial guess $x_1{\in}C$, to a common fixed point for such a continuous TAN family S in Banach spaces. Finally, some applications to a finite family of TAN self mappings are also added.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.2
• /
• pp.47-55
• /
• 2023

Recently, Sensor technologies are emerging to prevent traffic accidents and support safe driving in complex environments where human perception may be limited. The UWS is a technology that uses an ultrasonic sensor to detect objects at short distances. While it has the advantage of being simple to use, it also has the disadvantage of having a limited detection distance. The LDWS, on the other hand, is a technology that uses front image processing to detect lane departure and ensure the safety of the driving path. However, it may not be sufficient for determining the driving environment around the vehicle. To overcome these limitations, a system that utilizes FMCW radar is being used. The BSD radar system using FMCW continuously emits signals while driving, and the emitted signals bounce off nearby objects and return to the radar. The key technologies involved in designing the BSD radar system are tracking algorithms for detecting the surrounding situation of the vehicle. This paper presents a tracking algorithm for designing a BSD radar system, while explaining the principles of FMCW radar technology and signal types. Additionally, this paper presents the target tracking procedure and target filter to design an accurate tracking system and performance is verified through simulation.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.79-89
• /
• 2016

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.