• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1171-1172
• /
• 2023

자율주행 기술에서 스테레오 비전과 End-to-End Driving은 많이 사용되는 기술이며 본 연구에서는 이를 신호등 인식과 주행에 적용하였다. 신호등 인식은 좌우 카메라로부터 적색 원을 인식한 후 스테레오 비전을 통해 신호등과의 거리를 추정한다. 주행 시스템은 End-to-End Learning 기반으로 이루어지며, 출력값인 가변저항을 조향각으로 변환하여 제어할 수 있다. 또한 감마 보정을 통한 데이터 증강을 통해 빛에 대해 민감하지 않게 모델을 학습하였다. 추후 신호등 인식 시 HSV 필터가 빛에 민감한 점과 주행 시 가변저항 값이 일정하지 않은 점이 해결된다면 더욱 안정적인 시스템을 구축할 수 있을 것으로 기대된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.551-554
• /
• 2020

본 논문은 주행 로봇의 h/w에 관련된 연구로서, 기존의 험난한 지형을 극복하기 위해 1-자유도 반의 4-bar linkage 구조인 deformation wheel로 로봇 자체 지능을 통해 바퀴 변형을 수행한다. 바퀴변형을 통해 평지뿐만 아니라 비평지 지형도 극복하는 로봇을 제시한다. 또한, 로봇 몸체 중간에 관절로 다이나믹셀을 삽입해 deformation wheel로 극복하지 못하는 장애물을 관절이 로봇 body를 들어 올려줘서 장애물의 키기에 대한 관절의 각도 조절 방법에 대해 제시한다.

• Journal of Aerospace System Engineering
• /
• v.15 no.1
• /
• pp.72-79
• /
• 2021

In this study, as part of the development of an autonomous flying personal aircraft, an equivalent model of the main wing assembly of an Optionally Piloted Personal Air Vehicle (OPPAV) was developed. Reliability of the developed equivalent model was verified by eigenvalue analysis. The main wing assembly consisted of a main wing, an inboard pod, and an outboard pod. First, for developing an equivalent model of each component, components to produce the equivalent model were divided into several sections. Nodes were then created on the axis of the equivalent model at both ends of each section. In addition, static analysis with unit force and unit moment was performed to calculate the deformation or the amount of rotation at the node to be used in the equivalent model. Equivalent axial, bending, and torsional stiffness of each section were calculated by applying the beam theory. Once the equivalent stiffness of each section was calculated, information of a mass and moment of inertia for each section was entered by creating a lumped mass in the center of each section. An equivalent model was developed using beam element. Finally, the reliability of the developed equivalent model was verified by comparison with results of mode analysis of the fine model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.498-504
• /
• 2005

A fuzzy logic for collision avoidance of underwater robots is proposed in this paper. The VFF(Virtual Force Field) method, which is widely used in the field of mobile robots, is modified for application to the autonomous navigation of underwater robots. This Modified Virtual Force Field(MVFF) method using the fuzzy logic can be used in either track keeping or obstacle avoidance. Fuzzy logics are devised to handle various situations which can be faced during autonomous navigation of underwater robots. The obstacle avoidance algorithm has the ability to handle multiple static obstacles. Results of simulation show that the proposed method can be efficiently applied to obstacle avoidance of the underwater robots.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.8
• /
• pp.825-829
• /
• 2011

We performed a dynamic characterization of passive flow-rate regulators, which compensate for inlet pressure variation and maintain a constant flow rate for precise liquid control in microfluidic systems. To measure the flow rate for a short time, much less than the period of the dynamic inlet pressure, we use the particle image velocimetry (PIV) method. DI water containing fluorescent beads with a $0.7-{\mu}m$ diameter was supplied to the flow-rate regulators, and two successive images of the particles were taken by a pulse laser and a fluorescent microscope to measure the flow velocity. For a dynamic inlet pressure of frequency 60 Hz, the flow velocity was constant with an average of 0.194 ${\pm}$ 0.014 m/s as the inlet pressure varied between 20 kPa to 50 kPa. The flow-rate regulators provided a constant flow rate of $5.82{\pm}0.29\;{\mu}l/s$ in the frequency range of the inlet pressure from 1 Hz to 60 Hz.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.5
• /
• pp.163-181
• /
• 2023

Improving the detection performance of facilities corresponding to the sensors of autonomous vehicles helps driving safety. In the road and transportation field, research is being conducted to improve the detection performance of sensors by road infrastructure or facilities. As part of this on the development of autonomous driving support infrastructure, the shape of traffic cones and drums to ensure sufficient LiDAR detection performance even rainy conditions and maintain the line-of-sight guidance function in construction zones improvement effect. The principle was to increase reflection performance and ensure no significant difference in shape from existing facilities. Traffic cones were manufactured in square pyramid shapes instead of cones, and drums were manufactured in hexagonal and octagonal pillar shapes instead of cylinders. LiDAR detection data for the facility was confirmed on a clear day and with 20 mm/h and 40 mm/h rainfall. The detection performance of the square pyramid-shaped traffic cone and octagonal column-shaped drum was to the existing facility. On the other hand, deviations occurred due to repeated measurements, and significance could not be confirmed through statistical analysis. By reflecting these results, future studies will seek a form in which data can be obtained uniformly despite the diversity of measurement environments.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.9 no.4
• /
• pp.75-84
• /
• 2000

The main goal of our research is to compensate the milled surface errors induced by the tool deflection effects, which occur during the milling process. First, we predict cutting forces and tool deflection amount. Based on predicted deflection effects, we model milled surface shapes. We present a compensation methodology , which can generate a new tool trajectory, which is determined so as to compensate the milled surface errors. By considering manufacturing tolerance, tool path compensation is generalized. To validate the approaches proposed in this paper, we treat an illustrative example of profile milling process by using flat end mill. Simulation and experimental results are shown.

• Journal of Korea Game Society
• /
• v.9 no.1
• /
• pp.55-62
• /
• 2009

User participation, potentiality of modification, and interaction are some of significant features of game, and in the same sense, mod - user-created additions and modifications to an existing game - represents these aspects of game. Therefore, this study shows the possibility of customer involvement in the game development process through a case study of Half-life mod.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.440-442
• /
• 2019

메카넘 휠과 4족 및 6족의 다리를 갖는 탐사 및 재난 로봇을 바탕으로 휠과 6족이 함께 존재하는 탐사 및 재난 로봇이다. 평지나 완만한 길에선 메카넘 휠을 이용해 빠르게 굴러가는 모습을 볼 수 있다. 그리고 계단이나 산악지대 같은 경사가 높고 휠로 이동하기 힘든 공간은 부착된 6족 다리를 이용해 걸어 나가는 모습을 볼 수 있다. 이러한 휠과 6족이 결합된 탐사 및 재난 로봇은 사람의 조종 없이 영상처리와 여려가지 센서를 이용해 스스로 자율주행 하는 모습을 볼 수 있다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.199-202
• /
• 2022

왜곡된 영상에 강인한 물체 검출은 자율 주행과 같은 안전에 치명적인 실생활 응용 분야에서 핵심 문제로 다뤄지고 있다. 이러한 이유로, 영상 처리 및 컴퓨터 비전 분야에서 강인한 물체 검출에 대한 연구가 활발하게 진행되고 있다. 본 논문에서 우리는 왜곡된 영상이나 이미지에서도 일정한 성능의 물체 검출을 위한 연구 [4]를 benchmark 하여 다양한 방법으로 변형된 데이터셋을 통해 학습한 모델을 성능과 feature map 측면에서 분석해 봄으로써, 향후 강인한 물체 검출에 있어 효과적인 성능 향상을 위한 intuition 을 제공하고자 한다.