• 한국응급구조학회지
• /
• 제4권1호
• /
• pp.95-100
• /
• 2000

This is a study to evaluate the effects of the safety of ambulance driving and traffic accidents and to provide statistic information for the various factors to reduce the ambulance traffic accidents. The major instruments of this study were Korean Self-Analysis Driver Opinionnaire. This Questionnaire contains 8 items which measure drivers' opinions or attitudes: driving courtesy, emotion, traffic law, speed, vehicle condition, the use of drugs, high-risk behavior, human factors. The total of 145 divers were investigated ambulance drivers in Taejon City and others(6 City) from 2000. 5. July to 2000. 11. July. The data were analyzed by the path analysis - with SPSS and AMOS package program. The result are as follows : 1. It have suggested that risk factors of ambulance traffic accident much affected with emotion and speed control on safety ambulance driving(Y(Accident) = $0.88{\times}1$(Emotion Control) + $0.92{\times}2$(Speed) - $0.46{\times}3$(Traffic Law)+E). 2. It have suggested that risk factors of ambulance traffic accident much affected with emotion and speed control on safety ambulance driving(Y(Accident) = $0.398{\times}1$(Emotion Control) + $0.500{\times}2$(Speed) - $0.263{\times}3$(Traffic Law)+E) by coefficiecial structural analysis.

• 한국도시철도학회논문집
• /
• 제6권4호
• /
• pp.401-410
• /
• 2018

제4차 산업혁명은 철도분야에도 빠르게 자리 잡고 있다. 반면, 철도종사의 행동, 실제 근무여건 등을 고려한 과학적 분석 및 미흡한 대책으로 기관사의 인적오류로 인한 사고는 지속적으로 발생하고 있다. 이를 해결하기 위해 철도차량 운전면허 자격시험에 인적오류와 관련된 문제 출제, 시뮬레이터용 소프트웨어 개발 등 여러 가지 대책을 마련하고 있지만, 한계가 있는 것도 사실이다. 본 연구는 현재 가장 많은 인력을 양산하는 제2종 전기차량 운전면허를 전문가 인터뷰 및 설문조사에 대한 AHP 및 FGI 분석을 통해 제도 개선 방안을 연구하였다.

• Journal of Information Processing Systems
• /
• 제8권3호
• /
• pp.409-420
• /
• 2012

The decision-making by agents in games is commonly based on reinforcement learning. To improve the quality of agents, it is necessary to solve the problems of the time and state space that are required for learning. Such problems can be solved by Macro-Actions, which are defined and executed by a sequence of primitive actions. In this line of research, the learning time is reduced by cutting down the number of policy decisions by agents. Macro-Actions were originally defined as combinations of the same primitive actions. Based on studies that showed the generation of Macro-Actions by learning, Macro-Actions are now thought to consist of diverse kinds of primitive actions. However an enormous amount of learning time and state space are required to generate Macro-Actions. To resolve these issues, we can apply insights from studies on the learning of tasks through Programming by Demonstration (PbD) to generate Macro-Actions that reduce the learning time and state space. In this paper, we propose a method to define and execute Macro-Actions. Macro-Actions are learned from a human subject via PbD and a policy is learned by reinforcement learning. In an experiment, the proposed method was applied to a car simulation to verify the scalability of the proposed method. Data was collected from the driving control of a human subject, and then the Macro-Actions that are required for running a car were generated. Furthermore, the policy that is necessary for driving on a track was learned. The acquisition of Macro-Actions by PbD reduced the driving time by about 16% compared to the case in which Macro-Actions were directly defined by a human subject. In addition, the learning time was also reduced by a faster convergence of the optimum policies.

• 한국도로학회논문집
• /
• 제10권2호
• /
• pp.101-113
• /
• 2008

본 연구는 지방부 고속도로 상에서 유출지점용 도로전광표지의 적정 설치지점을 도출하기 위하여, 차량시뮬레이터 실험 자료를 변수로 운전자가 도로전광표지의 교통정보를 판독한 후 고소고도로 유출시설로 우회가 가능한지 여부를 판단할 수 있는 판별식(Discriminant Model for VMS installation : DMV식)을 개발하는 것이다. 연구결과로는 (첫째), 차량시뮬레이터 실험을 통하여 도로전광표지 설치지점을 변경시키면서 운전자의 인적, 행태적 자료를 수집하고, 이를 근거로 도로전광표지 설치지점에 따라 우회 가능성을 판단할 수 있는 DMV식을 개발하였다. (둘째) 우회가능성은 도로전광표지 설치지점이 멀수록, 운전자경력이 많을수록 고속도로 운전경험이 많을수록, 차량속도가 낮을수록 높아지는 것으로 나타났으며, 우회가능성에 큰 영향을 미치는 변수는 도로전광표지 설치지점, 고속도로 경험, 운전경력, 차량속도 순으로 나타났다. (셋째), 개발된 DMV식의 예측력을 검증한 결과, 개별 운전자가 도로전광표지의 교통정보를 판독하고 이에 대응하여 우회가능성을 정확하게 예측하였다. (넷째) DMV식을 이용하여 설치지점에 대한 우회가능성의 민감도분석을 수행한 결과, 지방부 고속도로에서 운전자의 85%이상이 도로전광표지로부터 교통정보를 판독하고 유출지점으로 진입하기 위해서는 유출지점으로부터 최소한 3.2km의 이격거리가 필요한 것으로 분석되었다. 이는 피실험자가 20, 30대 운전자로 대표성에 한계가 있음을 감안하더라도 유출지점으로부터 건설교통부 지침에서 제시한 3.0Km이상의 이격거리가 필요하다는 것을 시사한다 하겠다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제3권1호
• /
• pp.37-47
• /
• 1997

본 논문에서는 운전 시뮬레이션의 한 요소인 배경교통의 생성을 위한 알고리듬을 제안한다. 각 배경차량은 교통신호등, 도로 주변 표지판, 차선 그리고 수변의 차들을 인지하며 교통규칙에 따라 지능적으로 움직이는 하나의 autonomous agent로 모델된다. 본 논문의 알고리듬은 현재 상용자가 운전하는 차의 주변만을 국소적으로 고려하여, 시뮬레이션되는 가상 도로 환경의 크기에 관계없이 일정한 양의 계산에 소요되도록 하였다. 배경차량의 단조로운 행동양식을 피하기 위하여, 여러 종류의 확률분포함수를 통해 인위적인 다양성을 보이도록 하였다. 결과적으로 도출된 배경차량의 행동양식은 매우 사실적인 것으로 판명되었다. 실험에 의하면, 본 논문에서 제안한 알고리듬을 통해 생성된 배경차량의 움직임과 사람의 직접 조종하는 차량의 움직임은 쉽게 구별될 수 없었다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.696-701
• /
• 2004

A vehicle cruise control algorithm using an Interacting Multiple Model (IMM)-based Multi-Target Tracking (MTT) method has been presented in this paper. The vehicle cruise control algorithm consists of three parts; track estimator using IMM-Probabilistic Data Association Filter (PDAF), a primary target vehicle determination algorithm and a single-target adaptive cruise control algorithm. Three motion models; uniform motion, lane-change motion and acceleration motion, have been adopted to distinguish large lateral motions from longitudinal motions. The models have been validated using simulated and experimental data. The improvement in the state estimation performance when using three models is verified in target tracking simulations. The performance and safety benefits of a multi-model-based MTT-ACC system is investigated via simulations using real driving radar sensor data. These simulations show system response that is more realistic and reflective of actual human driving behavior.

• Journal of Mechanical Science and Technology
• /
• 제19권9호
• /
• pp.1742-1752
• /
• 2005

This paper presents a Multiple Target Tracking (MTT) Adaptive Cruise Control (ACC) system which consists of three parts; a multi-model-based multi-target state estimator, a primary vehicular target determination algorithm, and a single-target adaptive cruise control algorithm. Three motion models, which are validated using simulated and experimental data, are adopted to distinguish large lateral motions from longitudinally excited motions. The improvement in the state estimation performance when using three models is verified in target tracking simulations. However, the performance and safety benefits of a multi-model-based MTT-ACC system is investigated via simulations using real driving radar sensor data. The MTT-ACC system is tested under lane changing situations to examine how much the system performance is improved when multiple models are incorporated. Simulation results show system response that is more realistic and reflective of actual human driving behavior.

• 대한기계학회논문집A
• /
• 제29권1호
• /
• pp.139-144
• /
• 2005

A vehicle cruise control algorithm using an Interacting Multiple Model (IMM)-based Multi-Target Tracking (MTT) method has been presented in this paper. The vehicle cruise control algorithm consists of three parts; track estimator using IMM-Probabilistic Data Association Filter (PDAF), a primary target vehicle determination algorithm and a single-target adaptive cruise control algorithm. Three motion models; uniform motion, lane-change motion and acceleration motion. have been adopted to distinguish large lateral motions from longitudinal motions. The models have been validated using simulated and experimental data. The improvement in the state estimation performance when using three models is verified in target tracking simulations. The performance and safety benefits of a multi-model-based MTT-ACC system is investigated via simulations using real driving radar sensor data. These simulations show system response that is more realistic and reflective of actual human driving behavior.

• 대한인간공학회지
• /
• 제31권5호
• /
• pp.647-655
• /
• 2012

Objective: The purpose of this study was to confirm difference between angular foot movement time and existing foot Fitts' law predicting times, and to develop the angular foot Fitts' law in the foot tapping task. Background: Existing studies of foot Fitts' law focused on horizontal movement to predict the movement time. However, when driving a car, humans move their foot from the accelerator to the brake with a fixed heel. Therefore, we examined the experiment to measure angular foot movement time in reciprocal foot tapping task and compared to conventional foot Fitts' law predicting time. And, we developed the angular foot Fitts' law. Method: In this study, we compared the angular foot movement time in foot tapping task and the predicted time of four conventional linear foot Fitts' law models - Drury's foot Fitts' law, Drury's ballistic, Hoffmann's ballistic, Hoffmann's visually-controlled. 11 subjects participated in this experiment to get a movement time and three target degrees of 20, 40, and 60 were used. And, conventional models were calculated for the prediction time. To analyze the movement time, linear and arc distance between targets were used for variables of model. Finally, the angular foot Fitts' law was developed from experimental data. Results: The average movement times for each experiment were 412.2ms, 474.9ms, and 526.6ms for the 89mm, 172mm, and 253mm linear distance conditions. The results also showed significant differences in performance time between different angle level. However, all of conventional linear foot Fitts' laws ranged 135.6ms to 401.2ms. On the other hand, the angular foot Fitts' law predicted the angular movement time well. Conclusion: Conventional linear foot Fitts' laws were underestimated and have a limitation to predict the foot movement time in the real task related angular foot movement. Application: This study is useful when considering the human behavior of angular foot movement such as driving or foot input device.

• 한국안전학회지
• /
• 제28권2호
• /
• pp.78-83
• /
• 2013

The traffic culture index is used as a major index in evaluating the traffic safety services of local governments and also serve as important data for the planning and implementation of traffic safety services. However, as the traffic culture index gradually became a standard for comparison among local governments, in part, certain cases arose which questioned the grounds for selecting variables for the index and the validity of the index in terms of its influential relationship between evaluation items. This study analyzed the index's influential relationship by utilizing a PLS structural equation model based on the evaluation results of the 2011 traffic culture index. A variable-linking model was created which recognized the relativity taking into account of the indirect effects between latent variables and this model was proven to be a model suitable in explaining the traffic culture index with a 97.8% explanation power. It was found that traffic safety(0.530), driving behavior(0.527), pedestrian behavior(0.187) and vulnerable road users(0.147), in such order, had an effect on the traffic culture index. It was also found that human casualties due to traffic accidents under "traffic safety" and traffic light compliance rate under "driving behavior" had an important effect. The study showed that motor vehicle share in illegal parking in school zones did not have a valid explanation power regarding "vulnerable road users".