For real-time target classification, a study was conducted to improve the AI-based classification performance using RCS measurements that are vulnerable noise, but can be obtained quickly. To compensate for shortcomings of RCS, 1D CNN–GRU network with strengths in feature extraction and time-series processing considered. The experimentally changed designed fit characteristics. proposed compared analyzed CNN CNN–LSTM. had best high accuracy 99.50% complex situations, such as different missile...

Recently, autonomous driving technology has been in the spotlight. However, is still its infancy railway industry. In case of railways, there are fewer control elements than cars due to characteristics running on but a disadvantage that evasive maneuvers cannot be made event dangerous situation. addition, when braking, it decelerated quickly for weight body and safety passengers. tram, one systems, research already conducted how generate profile plans braking acceleration as base driving,...

In this paper, the dehazing algorithm is proposed using a one-channel grayscale depth image generated from LiDAR point cloud 2D projection image. image-based dehazing, estimation of scattering coefficient most important. Since coefficients are used to estimate transmission for optimal effective must be obtained depending on level haze generation. Thus, we estimated 100 synthetic images and represented distribution between dark channels. Moreover, through linear regression aforementioned...

Building damage due to various causes occurs frequently and has risk factors that can cause additional collapses. However, it is difficult accurately identify objects in complex structural sites because of inaccessible situations image noise. In conventional approaches, close-up images have been used detect segment such as cracks. this study, the method using a deep learning model proposed for rapid determination analysis multiple types, cracks concrete rubble, disaster sites. Through...

Dunaliella salina, a halophilic unicellular chlorophyte, produces bioactive compounds and biofuels applicable to various industries. Despite its industrial significance, comprehensive studies on the morphological, physiological, biochemical characteristics of genus remain challenging. In this study, we characterized an axenically isolated green alga from salt pond in Taean, Republic Korea, assessed industrially relevant traits. The morphological were typical D. molecular phylogenetic...

이 논문은 코로나19라는 초대형 재난 상황에서 우리나라 대학의 초기 대응을 살펴보되, 경북대학교를 중심으로 2020년 봄학기 비대면 수업의 학생만족도 등을 1차적으로 평가하였다. 우선, 정상수업이 이루어졌던 2019년과 견주어 핵심 교육수요층인 학생집단의 기말 강의평가를 비교하였다. 아울러, 감염병 확산 초기인 3월 말과 학기가 마무리된 7월 말을 이를 바탕으로 시점별로 달리 나타난 강의평가의 변화, 예컨대 불만족의 요인을 짚어보았다. 팬데믹과 강의 모두 미증유의 경험으로서, 본 연구는 어디까지나 거점국립대 하나를 사례로 살펴본 평가에 그쳐 연구 결과를 일반화할 수 없다는 한계를 지니지만, 가을학기 이후에도 계속될 가능성이 있는 효과성과 학생 만족도를 높일 정책 시사점을 찾고, 대학교육의 바이러스 대응에 대한 함의를 제공하려는 첫 번째 시도로서 의미가 있다. 더불어 교육 패러다임의 전환이 요청되는 시대에, 학생-교수-대학행정이 긴밀히 연결된 수업 및 성적관리는 학술적·실천적...

With the development of electric vehicles, indoor driving robots, wearable devices, etc., research on technology for recognizing one's current location continues to progress. Self-localization is an essential need autonomous and mainly requires global positioning system (GPS), thereby using technologies such as LiDAR, Wi-Fi, visible light communication (VLC). However, GPS greatly affected by its surrounding environment, it can make errors, especially in spaces where signals are blocked,...

The wheels of railway vehicles are paramount importance in relation to railroad operations and safety. Currently, the management vehicle is restricted post-event inspections whenever physical phenomena, such as abnormal vibrations noise, occur during operation vehicles. To address this issue, paper proposes a method for predicting abnormalities advance enhancing learning prediction performance machine algorithms. Data were collected Line 4 Busan Metro South Korea by directly attaching...

In this paper, a frequency tracking method based on cardinalized probability hypothesis density (CPHD) filter with cardinality compensation and fuzzy-c means (FCM) clustering is proposed to precisely estimate multiple interference frequencies in the received Global Navigation Satellite System (GNSS) signal. addition, an adaptive linear Kalman notch applied order mitigate GNSS signals by using results from CPHD filter. The refers number of frequencies, estimation accuracy has effect...

The prior covariance estimation method based on inverse intersection (ICI) is proposed to apply the particle flow filter. has better estimate performance and guarantees consistent results compared with previous works. ICI recently developed of ellipsoidal used get combined covariance. This integrates sample estimate, which unbiased but usually high variance, together a more structured typically biased target through fusion gains. For verifying algorithm, analysis simulations are performed....

The simulation of cohesive zone in the blast furnaces is conducted using 2-dimensional discrete element method. divided into two parts; (1) group flow coke and ore particles when reduced start to shrink disappear with melting, (2) biaxial compression test simulations breakable single particle receiving x y directional stresses which evaluated from simulation. In addition, a gasification model developed combined for process simulated by removing fixed amount.In simulation, have been found...

The purpose of this paper is to study the recognition ships and their structures improve safety drone operations engaged in shore-to-ship delivery service. This has developed a system that can distinguish between by using convolutional neural network (CNN). First, dataset Marine Traffic Management Net described CNN’s object sensing based on Detectron2 platform discussed. There will also be description experiment performance. In addition, been conducted actual operations—the first air service...

An adaptive Kalman filter-based frequency estimation algorithm is proposed to track the of a chirp-type global navigation satellite system (GNSS) interference. The unknown interference GNSS in received signal can be estimated by using samples time domain, but thus-obtained value include large amount error caused measurement noise and change rate In order reduce error, an fading filter that based on Fourier series applied algorithm. Furthermore, optimal order-reduction select appropriate for...

The fault tolerance estimation method is proposed to maintain reliable correspondences between sensor data and performance regardless of the number valid measurements. based on 1-point random sample consensus (RANSAC) unscented Kalman filter (UKF), inverse covariance intersection (ICI)-based fusion added update process in algorithm. To verify algorithm, two analyses are performed with respect degree measurement error reduction accuracy generated information. In addition, experiments...

We propose a sequential Monte Carlo-based cardinalized probability hypothesis density (SMC-CPHD) filter with adaptive survival for multiple frequency tracking to enhance the performance. The of particles in is adjusted using pre-designed exponential function related distribution estimated particle points. In order ensure whether proposed affects stability filter, error bounds prediction process are analyzed. Moreover, an inverse covariance intersection-based compensation method added...

Vehicle-to-everything (V2X) communication is an essential technology for autonomous driving above Level 4. This paper presents preview control of automotive actives suspension systems to improve ride comfort using vehicle-to-vehicle (V2V) communication. Generally, while preview-controlled active have even greater potential than feedback-controlled systems, their main challenge obtaining information the road profile ahead. A critical drawback "look-ahead" sensors increased risk incorrect...

A GNSS fault detection method using an adaptive fading Kalman filter is proposed for detecting faults such as step- and ramp-type bias error in pseudorange measurements. The factor of the used a parameter. In order to detect signals regardless type, based on single base station that has fixed location already known. simulations, different signal types are represented by ramp step pseudorange. change according quantitatively analyzed, threshold established analyzing covariance. addition,...