• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권7호
• /
• pp.2321-2338
• /
• 2021

Compared with vehicle trajectories, pedestrian trajectories have stronger degrees of freedom and complexity, which poses a higher challenge to trajectory prediction tasks. This paper designs a mode to divide the trajectory of pedestrians at a traffic intersection, which converts the trajectory regression problem into a trajectory classification problem. This paper builds a deep model for pedestrian trajectory prediction at intersections for the task of pedestrian short-term trajectory prediction. The model calculates the spatial correlation and temporal dependence of the trajectory. More importantly, it captures the interactive features among pedestrians through the Attention mechanism. In order to improve the training speed, the model is composed of pure convolutional networks. This design overcomes the single-step calculation mode of the traditional recurrent neural network. The experiment uses Vulnerable Road Users trajectory dataset for related modeling and evaluation work. Compared with the existing models of pedestrian trajectory prediction, the model proposed in this paper has advantages in terms of evaluation indicators, training speed and the number of model parameters.

• 드라이브 ㆍ 컨트롤
• /
• 제19권3호
• /
• pp.9-15
• /
• 2022

This paper proposes a method for constructing and verifying datasets used in deep learning technology, to prevent safety accidents in automated construction machinery or autonomous vehicles. Although open datasets for developing image recognition technologies are challenging to meet requirements desired by users, this study proposes the interface of virtual simulators to facilitate the creation of training datasets desired by users. The pixel-level training image dataset was verified by creating scenarios, including various road types and objects in a virtual environment. Detecting an object from an image may interfere with the accurate path determination due to occlusion areas covered by another object. Thus, we construct a database, for developing an occlusion area detection algorithm in a virtual environment. Additionally, we present the possibility of its use as a deep learning dataset to calculate a grid map, that enables path search considering occlusion areas. Custom datasets are built using the RDBMS system.

• 한국산업융합학회 논문집
• /
• 제25권6_2호
• /
• pp.1047-1053
• /
• 2022

The leaf spring for a truck absorbs shocks or vibrations from the road surface while driving with the elastic force of the material and prevents the shock from being transmitted to the vehicle body. It is subjected to cyclic stress, and fatigue fracture occurs frequently. This study analyzes fractured leaf spring from a 25 ton truck that has been operating for about a year. In the fractured portion, which is the origin of crack, inclusions were observed, and fatigue failure was caused by cyclic stress. In the stress calculation and FE analysis, the stress at the center of the leaf spring was obtained to be 54~65% of the yield strength of the base material and damaged material. It is most important to prevent the mixing of impurities in the steel manufacturing for leaf springs. The large stress portion of the leaf spring needs to introduce compressive residual stress by peening etc.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권8호
• /
• pp.1996-2015
• /
• 2023

High-definition (HD) maps can provide precise road information that enables an autonomous driving system to effectively navigate a vehicle. Recent research has focused on leveraging semantic segmentation to achieve automatic annotation of HD maps. However, the existing methods suffer from low recognition accuracy in automatic driving scenarios, leading to inefficient annotation processes. In this paper, we propose a novel semantic segmentation method for automatic HD map annotation. Our approach introduces a new encoder, known as the convolutional transformer hybrid encoder, to enhance the model's feature extraction capabilities. Additionally, we propose a multi-level fusion module that enables the model to aggregate different levels of detail and semantic information. Furthermore, we present a novel decoupled boundary joint decoder to improve the model's ability to handle the boundary between categories. To evaluate our method, we conducted experiments using the Bird's Eye View point cloud images dataset and Cityscapes dataset. Comparative analysis against stateof-the-art methods demonstrates that our model achieves the highest performance. Specifically, our model achieves an mIoU of 56.26%, surpassing the results of SegFormer with an mIoU of 1.47%. This innovative promises to significantly enhance the efficiency of HD map automatic annotation.

• 융합경영연구
• /
• 제11권2호
• /
• pp.13-22
• /
• 2023

Purpose: The purpose of this study is to examine the policy implications of establishing a compact city in Seoul, analyzing whether it is an appropriate and efficient eco-friendly housing supply alternative. Research design, data, and methodology: The analysis criteria include efficiency, safety, and comfort, with efficiency encompassing economic, energy, and public transport links' efficiency. Safety and comfort are aspects of eco-friendliness, housing safety, and improvement in living environments. Results: In terms of economic efficiency, compact cities are a less expensive option than purchasing land for housing construction. To increase energy efficiency, we plan to adopt eco-friendly energy sources. Transportation efficiency is high in locations near public transport stations. To enhance safety and comfort, we intend to create large-scale parks and forests in Seoul. To ensure residential safety, measures will be taken to reduce road vibration, vehicle noise, and scattering dust. Conclusions: Selecting an appropriate location that provides convenient public transportation is essential for creating a compact city for housing in a large city. Combining a compact and smart city is necessary, and implementing smart technologies is needed to prevent dust, noise, and vibrations, which are undesirable in a residential environment.

• 한국의류학회지
• /
• 제46권6호
• /
• pp.1128-1141
• /
• 2022

Due to increased interest in safety in sports, leisure, industries, and daily life; the demand for products that protect the head is increasing. As a preparatory study for the development of head protection for head injury prevention, this study analyzed patents and utility models related to head protection products such as industrial safety helmets, vehicle helmets, and sports protection gear. For this study, 368 patents and utility models for head protection products searched through WipsOn were selected and analyzed by application year, function, application, protection area, main material, and subject. From the analytic results of this study, the quantitative and qualitative flow and characteristics of developing technology related to head protection products were identified. Through the trend of current technology, it provided data to seek the development direction in the future. The significance of this study is to secure objective data to establish a road map for creating new Intellectual Property for head protection products.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.112-119
• /
• 2023

In urban environments, signals of Global Positioning System (GPS) can be blocked and reflected by tall buildings, large vehicles, and complex components of road network. Therefore, the performance of the positioning system using the GPS module in urban areas can be degraded due to the loss of GPS signals necessary for the position estimation. To deal with this issue, various localization schemes using inertial measurement unit (IMU) sensors, such as gyroscope and accelerometer, and Bayesian filters, such as Kalman filter (KF) and particle filter (PF), have been designed to enhance the performance of the GPS-based positioning system. Among Bayesian filters, the PF has been widely used for the target tracking and vehicle navigation, since it can provide superior performance in estimating the state of a dynamic system under nonlinear/non-Gaussian circumstance. This paper presents a positioning system that uses the double-stacked particle filter (DSPF) as well as the accelerometer, gyroscope, and GPS receiver on the smartphone to provide higher pedestrian positioning accuracy in urban environments. The DSPF employs a nonparametric technique (Parzen-window) to create the multimodal target distribution that approximates the posterior distribution. Experimental results show that the DSPF-based positioning system can provide the significant improvement of the pedestrian position estimation in urban environments.

• Smart Structures and Systems
• /
• 제31권3호
• /
• pp.301-309
• /
• 2023

Tilt is a key indicator of structural safety. Real-time monitoring of tilt responses helps to evaluate structural condition, enable cost-effective maintenance, and enhance lifetime resilience. This paper presents a prototype wireless sensing system for structural tilt measurement. Long range (LoRa) technology is adopted by the sensing system to offer long-range wireless communication with low power consumption. The sensor integrates a gyroscope and an accelerometer as the sensing module. Although tilt can be estimated from the gyroscope or the accelerometer measurements, these estimates suffer from either drift issue or high noise. To address this challenging issue and obtain more reliable tilt results, two sensor fusion algorithms, the complementary filter and the Kalman filter, are investigated to fully exploit the advantages of both gyroscope and accelerometer measurements. Numerical simulation is carried out to validate and compare the sensor fusion algorithms. Laboratory experiment is conducted on a simply supported beam under moving vehicle load to further investigate the performance of the proposed wireless tilt sensing system.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2008년도 추계학술발표대회
• /
• pp.1448-1451
• /
• 2008

VANET(Vehicle Ad-hoc Network)은 통신 기능을 지원하는 지능형 차량들로 이루어진 애드혹 네트워크 환경으로써 최근 들어서 그 연구가 매우 활발하게 진행되고 있는 분야이다. VANET은 원활한 교통 소통, 사고 방지 등 여러 가지 편리한 기능들을 제공하지만, 그 기반을 애드혹 네트워크에 두고 있기 때문에 애드혹 네트워크에서 발생할 수 있는 보안 문제가 그대로 발생하며, 또한 그 환경적 특성에 따라 추가적인 보안 요구사항 역시 존재한다. VANET 환경에서 가장 중요하게 요구되는 보안요소는 협력 운전(cooperative driving) 시 메시지 인증, 무결성, 부인방지 등과 특정 차량에 대한 추적을 할 수 없도록 하는 위치 프라이버시 보호이다. 그러나 이 가운데 사용자의 위치 프라이버시는 조건적으로 신뢰기관에 의한 추적 역시 가능해야 한다는 조건을 포함한다. 본 논문에서는 L. Martucci 등이 제안한 자체적으로 생성하고 인증하는 pseudonym 기법[1]을 이용하여 이러한 보안 요구사항들을 만족시키는 방법을 제안하고자 한다. 제안하는 기법은 최초 차량 등록 시 받은 비밀 인자를 이용하여 보안 통신을 하며 추가적으로 신뢰 기관으로의 통신이 필요하지 않는 효율적인 보안 기법이다. 또한 기존 연구에서 발생했던 저장 공간의 문제, RSU(Road Side Unit) 접근 문제 등을 해결한다.

• 한국입자에어로졸학회지
• /
• 제20권1호
• /
• pp.13-24
• /
• 2024

The concentrations of polycyclic aromatic hydrocarbons (PAHs) and elemental carbon (EC) in particulate matter, typical primary aerosols have decreased in Seoul between 2003 and 2018 (80% for PAHs and 85% for EC). The yearly mean benzo[a]pyrene (BaP) concentration has been lower than 1 ng/m3 since 2010-2011, the target value set by the European Union (EU) and China. A series of policies related to solid fuel and vehicle in South Korea and China should be effective in the reduction of the ambient PAHs and EC concentrations. But the emission data of PAHs and EC at both countries did not support that hypothesis. Possible causes are uncertainties in the emission inventories of primary carbonaceous aerosols in South Korea and China, although there may be a minor effect of the emissions from North Korea on the concentrations in Seoul. Thus the further policy directions against PAHs and EC such as improvements of emissions inventories and measurements, intensive regulation of non-road mobile sources and control of PAHs derivatives are discussed.