• Journal of Korean Society of Transportation
• /
• v.36 no.3
• /
• pp.169-183
• /
• 2018

Aggressive driving leads to a greater crash potential because it threatens surrounding vehicles. This study conducted traffic simulation experiments using driving behavior data obtained from multi-agent driving simulations. VISSIM traffic simulator and surrogate safety assessment model (SSAM) were used to identify the impacts of aggressive driving on traffic stream in terms of safety and operational efficiency. Market penetration rates (MPR) of aggressive driving vehicle, coupled with various traffic conditions, were taken into consideration in analyzing the impacts. As expected, it was identified that aggressive driving vehicles tended to deteriorate the traffic safety performance. From the perspective of operational efficiency, interesting results were observable. Under level of service (LOS) A, B, and C, it was observed that the average travel speed increased with greater MPRs. Conversely, the average travel speed decreased with under LOS D and E conditions. The outcome of this study would be effectively used for developing safety-related policies for reducing aggressive driving behavior.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.3
• /
• pp.73-82
• /
• 2017

As of December 2016, the number of registered automobiles in Korea exceeds 21million. As a result, greenhouse gas emission by transportation sector are increasing every year. It was concluded that the development of the driving strategy considering the driving behavior and the road conditions, which are known to affect the fuel efficiency and the greenhouse gas emissions, could be the most effective fuel economy improvement. Therefore, this study aims to develop a fuel efficient driving strategy in a complex linear section with uphill and curved sections. The road topography was designed according to 'Rules about the Road Structure & Facilities Standards'. Various scenarios were selected. After generating the speed profile, it was applied to the Comprehensive Modal Emission Model and fuel consumption was calculated. The scenarios with the lowest fuel consumption were selected. After that, the fuel consumption of the manual driver's driving record and the selected optimal driving strategy were compared and analyzed for verification. As a result of the analysis, the developed optimal driving strategy reduces fuel consumption by 21.2% on average compared to driving by manual drivers.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.1
• /
• pp.77-84
• /
• 2015

In 2011, greenhouse gas emissions of transport sector were 85.04 million $tonCO_2eq$ and road emissions accounted for 95% of total emissions in the transport sector. There are few innovative technologies to reduce greenhouse gas emissions aside from eco-driving education and public relation program. Therefore, this paper focused on analyzing optimal acceleration by certain road grades and suggested fuel-efficient driving method for various uphill sections. Scenarios were established by driving modes. Speed profiles were generated by scenarios and speed variations. Each speed profile applied to Comprehensive Modal Emission Model and then each fuel consumption was estimated. Driving mode and speed variation that minimized fuel consumption were driven according to grade percent and uphill distance. When driving in the eco-friendly mode of the driving and speed variation, reduction rate of fuel consumption was evaluated by comparison between eco-driving and cruise control mode. When a vehicle drove under eco-driving mode at 100kph, 90kph and 80kph on uphill road, fuel consumptions were reduced by 33.9%, 30.8% and 5.3%, respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.87-88
• /
• 2019

최근 기계학습 기술의 급속한 발전에 힘입어 자율주행을 위한 객체 인식 및 처리 기술 역시 비약적으로 발전하고 있다. 그러나 이러한 기계학습의 성능은 모델의 구조와 학습용 데이터의 품질에 영향을 받는다. 특히 주행환경을 잘 표현하는 학습데이터가 중요한데 전혀 새로운 도로, 주행환경, 장애물, 정적 혹은 동적 객체 등을 마주하면 정확도와 안정성에서 부정적인 영향을 받을 수 있는 것이다. 해외의 주행 데이터들에 크게 의존하고 있는 우리나라의 현실에 비춰 볼 때 국내 환경에 맞는 학습데이터를 쉽고 효율적으로 확보/관리/분석할 수 있게 하는 환경의 구축이 시급하다. 따라서 본 논문에서는 자율주행을 위한 기계학습 데이터를 효과적으로 관리하고 분석하기 위한 소프트웨어를 설계하고 개발하였다. 구체적으로는 수집된 영상들을 관리하는 기능, 영상에 존재하는 노이즈 제거 및 화질 개선 처리 기능, 학습 및 검증을 위한 메타 정보 태깅 기능, 태깅 정보의 통계적 분석 기능들을 포함한다. 개발한 소프트웨어는 우리나라에서 자체 촬영한 자율주행 학습 영상들에 대해 딥러닝 모델들을 학습하고 검증하는데 활용할 예정이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.755-756
• /
• 2023

본 논문에서는 비평탄 지형 주행이 가능한 이동형 로봇의 구조 설계를 효율적으로 하기 위한 방법을 제안하고 실제로 구현하였다. 다양한 보행과 계단과 같은 비평탄 지형에서의 보행 메커니즘에 적합한 4개의 바퀴 및 구동 모터의 위치와 효율적 구조를 목적에 맞게 최적화 설계하였으며, 소형 로봇 플랫폼의 동작에 필요한 저전력의 효율적 구조를 제안하였다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.15 no.4
• /
• pp.693-702
• /
• 2002

The aim of this study is to develop a 3-dimensional dynamic analysis model, capable of considering the interaction between vehicles and bridges more accurately. The dynamic analysis model is developed with the high-speed train (KTX) and a 2-span continuous prestressed concrete box girder bridge with a double track. The 20-car model is developed using the moving vehicle model for the regular trainset. Three-dimensional frame elements are used for the bridge model. Using the developed models, a dynamic behavior analysis program is coded. The analytical results are compared with the dynamic field test results and found to be valid to yield quite accurate dynamic responses. Based on the results of this study, the hybrid model, made up of the moving vehicle model for the heaviest power car and the moving force model for the other cars, is quite simple and effective without loosing the accuracy that much. Under the coincidence condition of two trains traveling with resonance velocity in the opposite directions, it is necessary to check not only the dynamic responses of the bridge with one-way traffic but those with two- way coincidence.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.6
• /
• pp.14-20
• /
• 2015

The Korea ministry of land, infrastructure and transport set the goal of cutting greenhouse gas emissions from the transport sector by 34.3% relative to the business as usual scenario by 2020. In order to achieve this goal, support is being given to education and information regarding eco-driving. As a practical measure, however, a vehicle control strategy for decreasing fuel consumptions and emissions is necessary. Therefore, this paper presents an optimized driving model in order to decrease fuel consumption. Scenarios were established by driving mode. The speed profile for each scenario applied to Comprehensive Modal Emission Model and then each fuel consumption was estimated. Scenarios and speed variation with the least fuel consumption were derived by comparing the fuel consumptions of scenarios. The optimized driving model was developed by the derived the results. The speed profiles of general driver were collected by field test. The speed profile of the developed model and the speed profile of general driver were compared and then fuel consumptions for each speed profile were analyzed. The fuel consumptions for optimized driving were decreased by an average of 11.8%.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.109-123
• /
• 2024

Research into vision-based end-to-end autonomous driving systems utilizing deep learning and reinforcement learning has been steadily increasing. These systems typically encode continuous and high-dimensional vehicle states, such as location, velocity, orientation, and sensor data, into latent features, which are then decoded into a vehicular control policy. The complexity of urban driving environments necessitates the use of state representation learning through networks like Variational Autoencoders (VAEs) or Convolutional Neural Networks (CNNs). This paper analyzes the impact of different image state encoding methods on reinforcement learning performance in autonomous driving. Experiments were conducted in the CARLA simulator using RGB images and semantically segmented images captured by the vehicle's front camera. These images were encoded using VAE and Vision Transformer (ViT) networks. The study examines how these networks influence the agents' learning outcomes and experimentally demonstrates the role of each state representation technique in enhancing the learning efficiency and decision- making capabilities of autonomous driving systems.

• Journal of Advanced Marine Engineering and Technology
• /
• v.36 no.4
• /
• pp.430-436
• /
• 2012

This study was conducted to measure battery's voltage drop in a compact electric vehicle to get driving performance in various driving situations. In the experiment, to evaluate the energy consumption and milage, system performance have measured with changing of the driving speed and the reduction of driving distance when the heater was operating. The battery of the car in this study is lead type storage battery. The driving velocity was changed from 10km/h to 50 km/h with 20km/h intervals and the operating step of the heating device. As results, the electronic consumption rate was maximum at 35 km/h of vehicle speed and if the driver turning the heater at maximum, capacity will lead to 35% of energy consumption increment.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.1
• /
• pp.172-191
• /
• 2023

This study was undertaken to strategize the mixed traffic operation of autonomous vehicles in the pilot zone. This was achieved by analyzing the changes expected when autonomous vehicles are mixed in the autonomous vehicle pilot zone. Although finding a safe and efficient traffic operation strategy is required for the pilot zone to serve as a test bed for autonomous vehicles, there is no available operation strategy based on the mixture of autonomous vehicles. In order to presents a traffic operation strategies for each period of autonomous vehicle introduction, traffic efficiency and safety analysis was performed according to the autonomous vehicle market percentage rate. Based on the analysis results, the introduction stage was divided into introductory stage, transition period, and stable period based on the autonomous vehicle market share of 30% and 70%. This study presents the following traffic operation strategies. Considering the traffic flow operation strategy, we suggest the advancement of the existing road infrastructure at the introductory stage, and operating an autonomous driving lane and the mileage system during the transition period. We also propose expanding the operation of autonomous driving lanes and easing the speed limit during the stable period. In the traffic safety strategy, we present a manual and legal system for responding to autonomous vehicle accidents in the introductory stage, an analysis of the causes of autonomous vehicle accidents and the implementation of preventive policies in the transition period, and the advancement of the autonomous system and the reinforcement of the security system during the stable period. Through the traffic operation strategy presented in this study, we foresee the possibility of preemptively responding to the changes of traffic flow and traffic safety expected due to the mixture of autonomous vehicles in the autonomous vehicle pilot zone in the future.