• ETRI Journal
• /
• v.43 no.4
• /
• pp.603-616
• /
• 2021

Over the last few years, autonomous vehicles have progressed very rapidly. The odometry technique that estimates displacement from consecutive sensor inputs is an essential technique for autonomous driving. In this article, we propose a fast, robust, and accurate odometry technique. The proposed technique is light detection and ranging (LiDAR)-based direct odometry, which uses a spherical range image (SRI) that projects a three-dimensional point cloud onto a two-dimensional spherical image plane. Direct odometry is developed in a vision-based method, and a fast execution speed can be expected. However, applying LiDAR data is difficult because of the sparsity. To solve this problem, we propose an SRI generation method and mathematical analysis, two key point sampling methods using SRI to increase precision and robustness, and a fast optimization method. The proposed technique was tested with the KITTI dataset and real environments. Evaluation results yielded a translation error of 0.69%, a rotation error of 0.0031°/m in the KITTI training dataset, and an execution time of 17 ms. The results demonstrated high precision comparable with state-of-the-art and remarkably higher speed than conventional techniques.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.2
• /
• pp.383-403
• /
• 2021

With the development of mobile edge computing (MEC), some late-model application technologies, such as self-driving, augmented reality (AR) and traffic perception, emerge as the times require. Nevertheless, the high-latency and low-reliability of the traditional cloud computing solutions are difficult to meet the requirement of growing smart cars (SCs) with computing-intensive applications. Hence, this paper studies an efficient offloading decision and resource allocation scheme in collaborative vehicular edge computing networks with multiple SCs and multiple MEC servers to reduce latency. To solve this problem with effect, we propose a context-aware offloading strategy based on differential evolution algorithm (DE) by considering vehicle mobility, roadside units (RSUs) coverage, vehicle priority. On this basis, an autoregressive integrated moving average (ARIMA) model is employed to predict idle computing resources according to the base station traffic in different periods. Simulation results demonstrate that the practical performance of the context-aware vehicular task offloading (CAVTO) optimization scheme could reduce the system delay significantly.

• Journal of the Korean Solar Energy Society
• /
• v.39 no.3
• /
• pp.1-7
• /
• 2019

As part of the government's energy policy, Zero Energy Building certification was launched on January of 2017. However, the three passive-housing rental housing projects are the only ZEB-certified detached housing since the certification's launch. The reason is that, in order for a detached housing to earn ZEB certification, it has to secure self-reliance in energy, and a photovoltaic system is the only viable renewable energy system. Therefore, conducting an analysis to optimize the photovoltaic system in an early design stage is strongly recommended. This study aimed to propose an optimal photovoltaic system design for a detached housing after analyzing through the ECO2 energy simulation of 44 cases, varying in a module type and efficiency, inclination and azimuth. As a result, 15 cases out of 44 cases were analyzed to satisfy ZEB evaluation criteria, and it is thought that these data could contribute greatly to the expansion of ZEB certification dissemination.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.1018-1021
• /
• 2009

Wireless Sensor Network collects a data from the specific area and the control is composed of small sensor nodes. Like this sensors to after that is established at the beginning are operated with the battery, the operational duration until several years must be continued from several months and will be able to apply the resources which is restricted in efficiently there must be. In this paper RHC (rounting history cache) applies in Directed Diffusion which apply a data central concept a reliability and an efficiency in data transfer course set. RHC algorithms which proposes each sensor node updated RHC of oneself with periodic and because storing the optimization course the course and, every event occurrence hour they reset the energy is wasted the fact that a reliability with minimization of duplication message improved.

• Phonetics and Speech Sciences
• /
• v.13 no.1
• /
• pp.45-51
• /
• 2021

End-to-end (E2E) automatic speech recognition (ASR) has achieved promising performance gains with the introduced self-attention network, Transformer. However, due to training time and the number of hyperparameters, finding the optimal hyperparameter set is computationally expensive. This paper investigates the impact of hyperparameters in the Transformer network to answer two questions: which hyperparameter plays a critical role in the task performance and training speed. The Transformer network for training has two encoder and decoder networks combined with Connectionist Temporal Classification (CTC). We have trained the model with Wall Street Journal (WSJ) SI-284 and tested on devl93 and eval92. Seventeen hyperparameters were selected from the ESPnet training configuration, and varying ranges of values were used for experiments. The result shows that "num blocks" and "linear units" hyperparameters in the encoder and decoder networks reduce Word Error Rate (WER) significantly. However, performance gain is more prominent when they are altered in the encoder network. Training duration also linearly increased as "num blocks" and "linear units" hyperparameters' values grow. Based on the experimental results, we collected the optimal values from each hyperparameter and reduced the WER up to 2.9/1.9 from dev93 and eval93 respectively.

• Steel and Composite Structures
• /
• v.48 no.6
• /
• pp.625-639
• /
• 2023

In this paper, a Taguchi-based finite element method (FEM) has been proposed and implemented to assess optimal design parameters for minimum static deflection in laminated composite plate. An orthodox mathematical model (based on higher-order shear deformation plate theory and Green-Lagrange geometrical nonlinearity) has been used to compute the nonlinear central deflection values of laminated composite plates according to Taguchi design of experiment via a self-developed MATLAB computer code. The lay-up scheme, aspect ratio, thickness ratio and the support conditions of the laminated composite plate structure were designated as the governable design parameters. Analysis of variance (ANOVA) is used to investigate the effect of diverse control factors on the nonlinear static responses. Moreover, regression model is developed for predicting the desired responses. The ANOVA revealed that the lay-up scheme alongside the support condition plays vital role in minimizing the central deflection values of laminated composite plate under uniformly distributed load. The conformity test results of Taguchi analysis are also in good agreement with the numerical experimentation results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.147-147
• /
• 2021

A partition of water distribution network (WDN) into district metered areas (DMAs) brings the efficiency and efficacy for water network operation and management (O&M), especially in monitoring pressure and leakage. Traditionally, the DMA configurations (i.e., number, shape, and size of DMAs) are permanent and cannot be changed occasionally. This leads to changes in water quality and reduced network redundancy lowering network resilience against abnormal conditions such as water demand variability and mechanical failures. This study proposes a framework to automatically divide a WDN into dynamic DMA configurations, in which the DMA layouts can self-adapt in response to abnormal scenarios. To that aim, a complex graph theory is adopted to sectorize a WDN into multiscale DMA layouts. Then, different failure-based scenarios are investigated on the existing DMA layouts. Here, an optimization-based model is proposed to convert existing DMA layouts into dynamic layouts by considering existing valves and possibly placing new valves. The objective is to minimize the alteration of flow paths (i.e., flow direction and velocity in the pipes) while preserving the hydraulic performance of the network. The proposed method is tested on a real complex WDN for demonstration and validation of the approach.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.2
• /
• pp.107-114
• /
• 2024

Mobile ad hoc networks represent self-configuring networks of mobile devices that communicate without relying on a fixed infrastructure. However, traditional routing protocols in such networks encounter challenges in selecting efficient and reliable routes due to dynamic nature of these networks caused by unpredictable mobility of nodes. This often results in a failure to meet the low-delay and low-energy consumption requirements crucial for such networks. In order to overcome such challenges, our paper introduces a novel multi-objective and adaptive routing scheme based on the Q-learning reinforcement learning algorithm. The proposed routing scheme dynamically adjusts itself based on measured network states, such as traffic congestion and mobility. The proposed approach utilizes Q-learning to select routes in a decentralized manner, considering factors like energy consumption, load balancing, and the selection of stable links. We present a formulation of the multi-objective optimization problem and discuss adaptive adjustments of the Q-learning parameters to handle the dynamic nature of the network. To speed up the learning process, our scheme incorporates informative shaped rewards, providing additional guidance to the learning agents for better solutions. Implemented on the widely-used AODV routing protocol, our proposed approaches demonstrate better performance in terms of energy efficiency and improved message delivery delay, even in highly dynamic network environments, when compared to the traditional AODV. These findings show the potential of leveraging reinforcement learning for efficient routing in ad hoc networks, making the way for future advancements in the field of mobile ad hoc networking.

• Nuclear Engineering and Technology
• /
• v.56 no.8
• /
• pp.3236-3241
• /
• 2024

Ion cyclotron resonance heating (ICRH) is an important auxiliary heating method applied to the China Fusion Engineering Test Reactor, which can effectively heat the ions and electrons in plasma. Owing to the harsh nuclear environment, neutronic analyses are required to verify tritium self-sufficiency and neutron-shielding requirements. In this study, a neutronics analysis of the ICRH antenna was conducted using the COre and System integrated engine for Reactor Monte Carlo (cosRMC) code to estimate the neutron flux, radiation damage, nuclear heating, gas generation rate of key components, and tritium breeding ratio (TBR), providing data support for the subsequent optimization of the shielding design. In addition, the neutron flux of the coils around the antenna was calculated to prevent the entry of neutrons that damage the magnetic field coils through the gaps between the port plugs and antenna, and the shielding effects of the port-plug antenna on the surrounding components were analyzed. Finally, the results obtained using the cosRMC and MCNP codes were compared, which and presented good agreement, thus verifying the reliability of the neutronic analysis using the cosRMC code.

• Membrane Journal
• /
• v.25 no.4
• /
• pp.332-342
• /
• 2015

In this study, to solve the major problem of reverse osmosis (RO) membrane, surface of reverse osmosis membrane was modified by silane-epoxy multi layer. Octyltrimethoxysilane (OcTES) was polymerized to membrane surface via cross-linking by Sol-gel method. n = 8 alkylgroup of OcTES formed the branch structure by self assembly. And for improve fouling resistance of RO membrane, Ether group of ethylene glycol diglycidyl ether (EGDE) was given to improve hydrophilicity of RO membrane surface by ring-opening. To analyze structure of RO membrane surface with FE-TEM and AFM. Membrane surface of the ridge and valley structure and the bridge structure was confirmed due to the multi-layer surface modification of OcTES and EGDE. And through the increase of the roughness, the branch structure was formed well on membrane surface. Through the XPS analysis was identified chemical structure of membrane surface. And confirmed that the hydrophilic surface modification is given to the surface of the film through a Contact angle analysis. In optimization of EGDE surface modification condition, was suitable 0.5 wt% EGDE concentraion and $70^{\circ}C$ ring-opening temperature. In result of fouling resistance test and MFI is SUL-H10, $PA-OcTES_{1.0}$, $PA-OcTES_{1.0}-EGDE_{0.5}$ 68.7, 60.4, 5.4 ($10E-8hr/mL^2$), multi-layer surface modified membrane improved fouling resistance.