• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.7
• /
• pp.1146-1151
• /
• 2003

In this paper, results of an analysis of driving behavior characteristics and a driver-adaptive control algorithm for adaptive cruise control systems have been described. The analysis has been performed based on real-world driving data. The vehicle longitudinal control algorithm developed in our previous research has been extended based on the analysis to incorporate the driving characteristics of the human drivers into the control algorithm and to achieve natural vehicle behavior of the adaptive cruise controlled vehicle that would feel comfortable to the human driver. A driving characteristic parameters estimation algorithm has been developed. The driving characteristics parameters of a human driver have been estimated during manual driving using the recursive least-square algorithm and then the estimated ones have been used in the controller adaptation. The vehicle following characteristics of the adaptive cruise control vehicles with and without the driving behavior parameter estimation algorithm have been compared to those of the manual driving. It has been shown that the vehicle following behavior of the controlled vehicle with the adaptive control algorithm is quite close to that of the human controlled vehicles. Therefore, it can be expected that the more natural and more comfortable vehicle behavior would be achieved by the use of the driver adaptive cruise control algorithm.

• Journal of the Ergonomics Society of Korea
• /
• v.29 no.1
• /
• pp.17-24
• /
• 2010

There has been recent interest in intelligent vehicle technologies, such as advanced driver assistance systems (ADASs) or in-vehicle information systems (IVISs) that offer a significant enhancement of safety and convenience to drivers and passengers. However, unsuitable design of HMI (Human Machine Interface) must increase driver distraction and workload, which in turn increase the chance of traffic accidents. Distraction in particular often occurs under a heavy driving workload due to multitasking with various electronic devices like a cell phone or a navigation system while driving. According to the 2005 road traffic accidents in Korea report published by the ROad Traffic Authority (ROTA), more than 60% of the traffic accidents are related to driver error caused by distraction. This paper suggests the structure of vehicle environment for real-time driving behavior monitoring system while driving which is can be used the driver workload management systems (DWMS). On-road experiment results showed the feasibility of the suggested vehicle environment for driving behavior monitoring system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.485-486
• /
• 2019

Simulated driving behavior is an important aspect of realistic simulation systems. To simulate natural driving behavior, this paper proposes an imitation learning method based on active learning that combines demonstration and experience. Driving demonstrations are collected from human drivers in a driving simulator. A driving behavior policy is learned from these demonstrations. The driving demonstration dataset is augmented with new demonstrations that the original demonstrations did not contain, in the form of behaviors from another driving behavior policy learned from experience. The final driving behavior policy is learned from an augmented demonstration dataset.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.5
• /
• pp.13-22
• /
• 2015

These days, human behavior (human factor), the main cause of traffic accidents, has drawn more attention. Research on driving behavior based on DBQ(Driver Behavior Questionnaire), the analysis tool of driving behavior, has been conducted actively. In domestic previous studies, their analysis subjects were limited to researchers or military officials, and their analysis methods were based on factor analysis and regression analysis. Therefore, this study tries to find the factors of general drivers' driving behavior that influence risk driving, and to analyze their influential relationship. Regarding study scope, general drivers with driving career were asked to answer DBQ questionnaire, and 300 effective samples were analyzed. In addition, previous studies were investigated to draw the three measurable attributes of DBQ-'Lapse, Mistake, and Violation'-as main factors of traffic accidents, and structural equation model was applied to design risk driving behavior model. To identify the difference between risk driving groups, this study made use of multiple group analysis. The analysis came to the following results: First, according to the examination of the hypothesis that 'Lapse, Mistake, and Violation factors will influence risk driving behavior', all factors were found to be statistically significant. Regarding their level of influence on risk driving behavior, Violation was 0.464, Lapse 0.383, and Mistake 0.158, and thus Violation was analyzed to be the most influential. Secondly, according to the examination of the hypothesis that 'the influence of Lapse, Mistake, and Violation factors on risk driving behavior will be different by risk group', the influence of Lapse on risk driving behavior was found to be different by risk group. It is expected that the study results will be used as a fundamental program to introduce traffic accident prevention program and education that takes violation and lapse into consideration.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.14 no.2
• /
• pp.32-38
• /
• 2006

A vehicle adaptive cruise control strategy based on human drivers' driving characteristics has been investigated. Human drivers driving characteristics have been analyzed using vehicle test data obtained from 125 participants. The control algorithm has been designed to incorporate the driving characteristics of the human drivers and to achieve natural vehicle behavior of the controlled vehicle that would reduce the workload of the human driver. Vehicle following characteristics of the cruise controlled vehicle have been compared to real-world driving radar sensor data of human drivers using a validated vehicle simulator. and compare nominal cruise control and adaptive cruise control.

• Journal of the Korean Society of Safety
• /
• v.22 no.1 s.79
• /
• pp.61-66
• /
• 2007

Car crashes are the leading cause of death for persons of every age. Specially, human-related factor has been known to be the primary causal factor of such crashes than vehicle-and environmental-related factors. There are various studies to analyze driver's behavior and characteristics in driving for reducing the car crashes in many areas of car engineering, psychology, human factor, etc. However, there are almost no studies which analyze mainly the human errors in driving and estimate their probabilities in terms of human reliability analysis. This study estimates the probability of human error in driving, i.e. driver error probability. First, fifty driver errors are investigated through DBQ (Driver Behavior Questionnaire) revision and the error likelihoods in driving are collected which are judged by skillful drivers using revised DBQ. Next, these likelihoods are converted into driver error probabilities using the results that verbal probabilistic expressions are changed into quantitative probabilities. Using these probabilities we can improve the warning effects on drivers by indicating their driving error likelihoods quantitatively. We can also expect the reduction effects of car accident through controlling especially dangerous error groups which have higher probabilities. Like these, the results of this study can be used as the primary materials of safety education on drivers.

• Journal of Mechanical Science and Technology
• /
• v.19 no.4
• /
• pp.993-1000
• /
• 2005

A vehicle cruise control strategy designed based on human drivers driving characteristics has been investigated. Human drivers driving patterns have been investigated using vehicle driving test data obtained from 125 participants. The control algorithm has been designed to incorporate the driving characteristics of the human drivers and to achieve natural vehicle behavior of the controlled vehicle that would feel comfortable to the human driver. Vehicle following charac­teristics of the cruise controlled vehicle have been investigated using real-world vehicle driving test data and a validated simulation package.

• International Journal of Automotive Technology
• /
• v.7 no.5
• /
• pp.619-624
• /
• 2006

This paper presents design of a vehicle stop-and-go cruise control strategy based on analyzed results of the manual driving data. Human drivers driving characteristics have been investigated using vehicle driving data obtained from 100 participants on low speed urban traffic ways. The control algorithm has been designed to incorporate the driving characteristics of the human drivers and to achieve natural vehicle behavior of the controlled vehicle that would feel comfortable to the human driver under low speed stop-and-go driving conditions. Vehicle following characteristics of the cruise controlled vehicle have been investigated using a validated vehicle simulator and real driving radar sensor data.

• Journal of Korean Society of Transportation
• /
• v.31 no.6
• /
• pp.34-42
• /
• 2013

The reasons of traffic accidents can be classified into the vehicle, road environment and human factors. In development of vehicle and road technologies, the human factor is the most important. The emotion research recently carried out showed that emotions play an important role in human judgment and decision making. They are also expected to partly determine driving behavior. The purpose of this study was to investigate the effects of drivers' emotions on their driving behavior. The results of this study showed that positive emotions do not affect driving behavior. As for negative emotions, however, safe driving(SD) and comfort driving(CD) factors have a negative correlation, and violence driving(VD) and regulation violation(RO) factors have a positive correlation. This study is significant as a basic study for establishing and determining a policy for reducing traffic accidents and for educating drivers.

• Journal of Auto-vehicle Safety Association
• /
• v.15 no.2
• /
• pp.21-27
• /
• 2023

This paper presents a simulation tool for developing and evaluating automated driving systems' lane change algorithm in urban congested traffic. The behavior of surrounding vehicles was modeled based on driver driving data measured in urban congested traffic. Surrounding vehicles are divided into aggressive vehicles and non-aggressive vehicles. The degree of aggressiveness is determined according to the lateral position to initiate interaction with the vehicle in the next lane. In addition, the desired velocity and desired time gap of each vehicle are all randomly assigned. The simulation was conducted by reflecting the cognitive limitations and control performance of the autonomous vehicle. It was possible to confirm the change in the lane change performance according to the variation of the lane change decision algorithm.