• Annual Conference of KIPS
• /
• 2015.10a
• /
• pp.538-540
• /
• 2015

안전 거리 미확보는 교통 사고 원인 중 3위에 이르는 큰 위험 요소이다. 이러한 차간거리 미확보로 인한 사고는 조금만 빨리 대응하여도 사전에 방지할 수 있다. 따라서 주행 중 실시간으로 차간 거리를 추정하여 제공하는 시스템이 필요하다. 본 논문에서는 안드로이드 기반의 실시간 촬영 영상에서 차량의 에지와 후미등을 이용하여 차량을 검출하고, 검출된 차량의 폭을 이용하여 차간 거리를 추정하여 제공하는 시스템을 제안하고, 구현 결과를 제시한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.139-141
• /
• 2019

This paper proposes the application of LoRa network, which has been under various research recently, to use control system for the transfer of Unmanned Vehicle. LoRa( Long Range) has a wide coverage of up to30 km, and is characterized by large-scale low-power long-distance radio communication. LoRa has already been studied in various industries. It is also proposed to study various functions by utilizing LoRa network and applying it to the control system of the uninhabited entity.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.3
• /
• pp.13-19
• /
• 2021

Fog is a phenomenon caused by condensation of water vapor in the atmosphere, which is when very fine drops of water float in the atmosphere and the distance of visible is less than 1km. Fog dispersion technology is a technology that removing or weakening fog by using artificial methods to reduce damage caused by fog. It is applied differently depending on the temperature of fog generation rather than the cause of fog. This study conducted an experimental study on the fog dispersion mechanism in order to minimize damage caused by fog on the road, and studied two methods of over-cooling dispersion using solid-carbon-dioxide as a dissipated particle and dissipating fog particles through thermal acoustic waves. As a result the two methods proved experimentally that were capable of dissipating fog.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.728-734
• /
• 2021

In recent years, there has been a strong interest in the end-to-end autonomous driving based on deep reinforcement learning. In this paper, we present a reward function of latent SAC deep reinforcement learning to improve the longitudinal driving performance of an agent vehicle. While the existing reward function significantly degrades the driving safety and efficiency, the proposed reward function is shown to maintain an appropriate headway distance while avoiding the front vehicle collision.

• Journal of Auto-vehicle Safety Association
• /
• v.6 no.1
• /
• pp.16-21
• /
• 2014

Recently Korea and many countries are legislated for pedestrian protection as following the GTR(Global Technical Regulation). Most NCAP organization have been applying pedestrian protection results in their own rating system in individual or overall rating. Euro NCAP agency has been introduced it first. From the beginning, Euro NCAP had assessed occupant and pedestrian protection. It has become to strengthen a pedestrian protection protocol every time published. Korea NCAP also has been rating pedestrian protection with overall rating system. This paper proposed to study new pedestrian protection protocol especially upper leg in order to find injury response based on physical test. On Euro NCAP, the upper leg injury are assessed two kind of value which are bending moment and force.

• Korean Management Science Review
• /
• v.28 no.3
• /
• pp.1-13
• /
• 2011

Hub-and-spoke transportation network is a powerful and useful network structure that takes full advantage of economies of scale on routes between hubs. In recent studies, the network structure is extended to hybrid hub-andspoke that allows direct transportation between spokes. In this study, we considered more extended network structure which is called hybrid multiple hub-and-spoke that has multiple hubs and allows direct transportation between spokes. We developed a mathematical optimization model for automotive service parts transportation planning under hybrid multiple hub-and-spoke network structure. The model suggests a long-term transportation route planning and a short-term vehicle assignment planning. The model is verified by simulation and validated in real world application to Hyundai Mobis automotive service parts transportation planning. From the simulation result, the model reduced the transportation cost about 24.7%, the total distance about 6.8% and the CO2 emissions about 8.8%. In real world application for 6 months from July to December 2010, the model reduced the transportation cost about 9.1% by changing the long-term transportation route without daily vehicle assignment planning.

• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.8
• /
• pp.702-708
• /
• 2013

Underwater TRN (Underwater Terrain Referenced Navigation) estimates an underwater vehicle state by measuring a distance between the vehicle and undersea terrain, and comparing it with the known terrain database. TRN belongs to absolute navigation methods, which are used to compensate a drift error of dead reckoning measurements such as IMU (Inertial Measurement Unit) or DVL (Doppler Velocity Log). However, underwater TRN is different to other absolute methods such as USBL (Ultra-Short Baseline) and LBL (Long Baseline), because TRN is independent of the external environment. As a magnetic-field-based navigation, TRN is a kind of geophysical navigation. This paper develops an EKF (Extended Kalman Filter) formulation for underwater TRN. A filter propagation part is composed by an inertial navigation system, and a filter update is executed with echo-sounder measurement. For large-initial-error cases, an adaptive EKF approach is also presented, to keep the filter be stable. At the end, simulation studies are given to verify the performance of the proposed TRN filter. With simplified sensor and terrain database models, the simulation results show that the underwater TRN could support conventional underwater navigation methods.

• Spatial Information Research
• /
• v.13 no.3 s.34
• /
• pp.211-220
• /
• 2005

To understand present road state quickly and correctly, The Ministry of Construction and Transportation pushing on with their plan that they offer digital images to user with position information in the HMS(Highway Management System). For that plan, the Korea Institute of Construction Technology developed a highway photologging vehicle which could acquire two high resolution color CCD images with fixed distance on the way to run. In this paper, especially, development of image aquisition S/W and synchronization device are described. And also their performance and trajectory accuracy are investigated together. As a result, it could be hon that the newly developed highway photologging vehicle is suitable to the road image acquisition work far the HMS.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.20 no.3
• /
• pp.13-19
• /
• 2012

The fatalities rate for passenger vehicles, vans, and commercial vehicles is 1.23, 1.90 and 2.46 deaths per 10,000 registered vehicles, respectively. This shows that vans and commercial vehicles are vulnerable compare to passenger vehicles. To evaluate the crashworthiness of van and Light Truck Vehicle(LTV), we carried out frontal offset crash test at 64km/h, 40% overlap as per IIHS(Insurance Institute for Highway Safety). The test result show that LTV is very poor to protect occupant at frontal crash cause there is no safety system such as airbag and pretensioner and front end length(distance from front bumper to steering wheel) is short. One of the van rated as the lowest rating even it is equipped with airbag, cause its safety cage was collapsed during the test. This result shows that the structural integrity is very important in terms of occupant protection.

• Journal of Ocean Engineering and Technology
• /
• v.30 no.3
• /
• pp.208-213
• /
• 2016

This paper proposes a potential field-based guidance law for docking a USV (unmanned surface vehicle). In most cases, a USV without side thrusters is an under-actuated system. Thus, there are undockable regions near docking stations where a USV cannot dock to a docking station without causing a collision or backward motion. This paper suggest a guidance law that prevents a USV from enter such a region by decreasing the lateral error to the docking station at the initial stage of the docking process. A Monte-carlo simulation was performed to validate the performance of the proposed method. The proposed method was compared to conventional guidance laws such as pure pursuit guidance and pure/lead pursuit guidance. As a result, the collision angle and lateral distance error of proposed method tended to have lower values compared to conventional methods.