• ETRI Journal
• /
• 제45권5호
• /
• pp.795-810
• /
• 2023

Unmanned aerial vehicles (UAV) and ground vehicles (UGV) require advanced video analytics for various tasks, such as moving object detection and segmentation; this has led to increasing demands for these methods. We propose a zero-shot video object segmentation method specifically designed for UAV and UGV applications that focuses on the discovery of moving objects in challenging scenarios. This method employs a background memory model that enables training from sparse annotations along the time axis, utilizing temporal modeling of the background to detect moving objects effectively. The proposed method addresses the limitations of the existing state-of-the-art methods for detecting salient objects within images, regardless of their movements. In particular, our method achieved mean J and F values of 82.7 and 81.2 on the DAVIS'16, respectively. We also conducted extensive ablation studies that highlighted the contributions of various input compositions and combinations of datasets used for training. In future developments, we will integrate the proposed method with additional systems, such as tracking and obstacle avoidance functionalities.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2011년도 정기 학술대회
• /
• pp.324-327
• /
• 2011

The design requirement for ground mounted sign structures are fairly well defined in the AASHTO Standard Specifications for Structural Supports for Highway Signs, Luminaries, and Traffic Signals and consists of applying an equivalent pseudo-dynamic loading to account for the dynamic effects of wind loads and ignores the dynamic effect due to moving vehicle loads. This design approach, however, should not be applied to the design of bridge mounted sign structures because ignoring the dynamic effects of the moving vehicle loads may produce non-conservative results, since the stiffness of the bridge structure can greatly influence the behavior. Not enough information is available in the literatures which provide guide lines to include the influence of moving vehicles in the design of the bridge mounted sign structures. This paper describes a theoretical methodology, Bridge-Vehicle Interaction Element, which can be utilized to account for the dynamic effect of moving vehicles. A case study is also included where this methodology was successfully applied. It was concluded that the bridge-vehicle interaction finite element developed can provide a more accurate representation of the behavior of bridge mounted sign structures. The result of these analysis enabled development of simple and effective retrofitting scheme for the existing support system of bridge-mounted-structure.

• Structural Engineering and Mechanics
• /
• 제46권5호
• /
• pp.713-734
• /
• 2013

Bridges are vital components of the railroads. High speed of travel, the periodic and oscillatory nature of the loads and the comparable vehicle bridge weight ratio distinguish the railway bridges from the road bridges. The close proximity between estimations by some numerical methods and the measured data for the bridge-vehicle dynamic response under the moving load conditions has boosted the confidence in the numerical analyses. However, there is hardly any report regarding the responses of the railway bridges under the effect of the trains entering from the opposite directions while running at unequal speed and having dissimilar geometries. It is the purpose of this article to present an analytical method for the dynamic analysis of the railway bridges under the influence of two opposing series of moving loads. The bridge structural damping and many modes of vibrations are included. The concept of modal superposition is used to solve for the system motion equations. The method of solution is indeed a computer assisted analytical solution. It solves for the system motion equations and gives output in terms of the bridge deflection. Some case studies are also considered for the validation of the proposed method. Furthermore, the effects of varying some parameters such as the distance between the bogies, and the bogie wheelset distance are studied. Also, the conditions of resonance and cancellation in the dynamic response for a variety of vehicle-bridge specifications are investigated.

• 대한임베디드공학회논문지
• /
• 제13권2호
• /
• pp.65-71
• /
• 2018

In order to make the underwater vehicle carry out the mission in a submarine environment, it is needed to plan a safe and efficient route to a given destination and prevent the autonomous submersible from colliding with obstacles while moving along the planned route. The function of collision avoidance makes the travel distance of the autonomous submersible longer. Moreover, it should move slowly near to obstacles against their moving disturbance. As a result, this invokes the degradation of the navigation efficiency in the process of collision avoidance. The side effect of the collision avoidance is not ignorable in the case of high congested environments such as the coast with many obstacles. In this paper, we suggest a path planning method which provides the route with minimum travel time considering collision avoidance in congested environment. For the purpose, we define environmental congestion map related to geometric information and obstacles. And we propose a method to consider the moving cost in the RRT scheme that provides the existing minimum distance path. We verified that the efficiency of our algorithm with simulation experiments.

• Smart Structures and Systems
• /
• 제21권4호
• /
• pp.449-467
• /
• 2018

In this paper, a theoretical and numerical study on bridge simultaneous damage detection procedure for identifying both the system parameters and input excitation mass, are presented. This method is called 'Adjoint Variable Method' which is an iterative gradient-based model updating method based on the dynamic response sensitivity. The main advantage of proposed method is inclusion of an analytical method to augment the accuracy and speed of the solution. Moving mass is a model which takes into account the inertia effects of the vehicle. This interaction model is a time varying system and proposed method is capable of detecting damage in this variable system. Robustness of proposed method is illustrated by correctly detection of the location and extension of predetermined single, multiple and random damages in all ranges of speed and mass ratio of moving vehicle. A comparison study of common sensitivity and proposed method confirms its efficiency and performance improvement in sensitivity-based damage detection methods. Various sources of errors including the effects of measurement noise and initial assumption error in stability of method are also discussed.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.932-937
• /
• 2003

In this paper, we will explain about the unmanned vehicle control and modeling for combined obstacle avoidance and lane tracking. First, obstacle avoidance is considered as one of the important technologies in the unmanned vehicle. It is consisted by two parts: the first part includes the longitudinal control system for the acceleration and deceleration and the second part is the lateral control system for the steering control. Each system uses to the obstacle avoidance during the vehicle moving. Therefore, we propose the method of vehicle control, modeling and obstacle avoidance. Second, we describe a method of lane tracking by means of vision system. It is important in the unmanned vehicle and mobile robot system. In this paper, we deal with lane tracking and image processing method and it is including lane detection method, image processing algorithm and filtering method.

• 대한전자공학회논문지SP
• /
• 제45권4호
• /
• pp.10-21
• /
• 2008

본 논문은 실시간 감시 시스템을 위한 웨이블릿(wavelet) 기반의 신경망과 불변 모멘트를 이용한 이동물체 인식과 추적 방법을 제안한다. 제안한 방법의 첫 번째인 움직임 후보영역 검출 단계에서는 연속된 두 프레임간의 차영상 분석 방법을 기반으로 하여 물체의 움직임에 의해 화소값 변화가 발생한 후보영역을 검출한다. 두 번째인 물체 인식 단계에서는 검출된 후보영역에 웨이블릿 신경망(wavelet neural network: WNN) 기반의 인식 방법을 사용하여 추적하고자하는 물체가 포함되어 있는지를 판별한다. 세 번째인 물체 추적 단계에서는 인식된 물체에 웨이블릿 불변 모멘트(invariant moments) 기반의 매칭 방법을 사용하여 인식된 이동 물체를 추적한다. 영상내에서 이동물체를 검출하기 위해 본 논문에서는 이전 영상과 현재 영상간의 화소밝기 차이에서 적응적 임계값(adaptive threholding)을 사용하여 주위 환경 변화에 강인한 이동물체 검출이 가능하였다. 또한 물체의 인식과 추적을 위해 웨이블릿 특징값을 사용함으로써, 계산 시간의 감소와 영상의 잡음에 의한 영향을 최소화시킬 수 있을 뿐만 아니라, 물체 인식 정확도가 향상되었다. 제안한 방법을 일반 도로에서 획득한 영상에서 실험한 결과, 자동차 검출율은 92.8%, 프레임당 처리 시간은 0.24초이다. 이것을 통해 제안한 방법은 실시간 지능형 교통 감시 시스템에 유용하게 적용될 수 있음을 알 수 있다.

• Smart Structures and Systems
• /
• 제13권5호
• /
• pp.775-796
• /
• 2014

This study presents a vibration-based health monitoring strategy for short span bridges utilizing an inspection vehicle. How to screen the health condition of short span bridges in terms of a drive-by bridge inspection is described. Feasibility of the drive-by bridge inspection is investigated through a scaled laboratory moving vehicle experiment. The feasibility of using an instrumented vehicle to detect the natural frequency and changes in structural damping of a model bridge was observed. Observations also demonstrated the possibility of diagnosis of bridges by comparing patterns of identified bridge dynamic parameters through periodical monitoring. It was confirmed that the moving vehicle method identifies the damage location and severity well.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2005년도 추계학술대회 논문집
• /
• pp.1261-1267
• /
• 2005

At this paper Dynamic respones of bridges for the Korean high-speed railway are analyzed by a modal analysis. To control vibration of bridges, Tuned Mass Damper(TMD) that is passive type control device is used. Opimize and prove it. Newmark method is used for a numerical analysis. In case of vehicle is modeled for moving mass that considers the effects of the moving. Also this paper is assumped as the simple supported Bernoulli-Euler beam and considered two dimensional Interaction motion between vehicle and bridge.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 추계학술대회
• /
• pp.95-97
• /
• 2010

최근 차량에는 부착된 많은 전장기기를 제어하는 CAN 네트워크 기술과 더불어 엔터테인먼트 서비스를 제공하는 MOST 네트워크 기술이 도입되었다. MOST 네트워크에는 CD-ROM(DVD), AMP, VIDEO CAMERA, VIDEO DISPLAY, GPS NAVIGATION 등과 같은 많은 장치들이 연결되어 동작한다. 본 논문에서는 이런 MOST네트워크에 연동되는 CAMERA의 입력 영상을 차량의 이동 방향 결정에 이용하고자 한다. GPS로부터 위치정보를 받는다 하더라도 특정 구역에서는 평행한 도로구조로 인해 차량이 어느 방향으로 이동했는지 즉시 판단하기 어려운 경우가 발생한다. 이때 구축된 영상이미지와 CAMERA 영상을 실시간 매칭 처리하여 차량의 이동 방향을 결정하는 기법을 설계하고 구현하고자 한다.