• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.6
• /
• pp.759-764
• /
• 1998

We propose a novel algorithm capable of recognizing the road lane by image processing. Considering the fact that the direction and location of road lane are maintained similarly in successive images we formulate a function to represent the property. However, as noises play the role of making a lot of similar patterns appear and disappear in the road image, keeping of robustness in the lane detection has been known a difficult work. To overcome this problem, we introduce the following three ideas: 1) design of a function based on an edge direction and magnitude, 2) construction of a recursive filter to estimate the function recursively for successive images, 3) principal axis-based line fitting. These concepts enhance the adaptability to cope with the random environment of traffic scene and eventually lead to the reliable detection of a road lane.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.6D
• /
• pp.785-791
• /
• 2011

Two-lane and two-way traffic flow shows various dynamic relationships according to the behaviors of low-speed vehicle and overtaking. And it is essential to develop a vehicle model which simultaneously explains the behaviors of low-speed vehicle and overtaking using opposite lane in order to microscopically analyze various two-lane and two-way traffic flows by traffic flow simulation. In Korea, some studies for car-following and lane-changing models for freeway or signalized road have been reported, but few researches for the development of vehicle model for two-lane and two-way highway have been done. Hence, a microscopic two-lane and two-way vehicle model was, in this study, developed with the consideration of overtaking process and is based on CA (Cellular Automata) which is one of discrete time-space models. The developed model is parallel combined with an adjusted CA car-following model and an overtaking model. The results of experimental simulation showed that the car-following model explained the various macroscopic relationships of traffic flow and overtaking model reasonably generated the various behaviors of macroscopic traffic flows under the conditions of both opposite traffic flow and stochastic parameter to consider overtaking. The vehicle model presented in this study is expected to be used for the simulation of more various two-lane, two-way traffic flows.

• International Journal of Highway Engineering
• /
• v.19 no.6
• /
• pp.183-190
• /
• 2017

PURPOSES : The objective of this study is to establish the traffic volume-based warrants of right-turn lanes at unsignalized intersections and to introduce a risk probability methodology based on the warrants. METHODS : In this study, a risk probability of a potential rear-end collision is applied between a right-turn vehicle and the immediately following through vehicle. Using the shifted negative exponential model and the compound probability theorem, the risk probability can be expressed as the function of directional volumes and the percentage of right-turns for a two-lane and four-lane highway, respectively. RESULTS : Based on the risk probablity, guidelines for installing right-turn lanes on two-lane and four-lane highways were developed. The risk probability also showed rationality by comparing with right-turn same-direction conflicts observed in-situ. CONCLUSIONS : The results of our study define the total approaching volumes to encourage a right-turn lane as a function of operating speed, percentage of right-turn, and number of lanes.

• International Journal of Highway Engineering
• /
• v.14 no.6
• /
• pp.131-138
• /
• 2012

PURPOSES : Two-lane highways have one lane in each direction, and lane changing and passing maneuvers take place in the opposing lane depending on the availability of passing sight distance. 2001 Korea Highway Capacity Manual (KHCM) is classified into two classes of two-lane highways (Type I, II), and average travel speed and time-delayed rate are used as measures of effectiveness (MOEs). However, since existing two-lane highways have both uninterrupted and interrupted traffic flow-system elements, a variety of free-flow speeds exhibits in two-lane highways. In addition, it is necessary to check if the linear-relationship between volumes and time-delayed rate is appropriate. Then, this study is to reestablish the relationship between average travel speed, time-delayed rate, and flow. METHODS : TWOPAS model was selected to conduct this study, and the free-flow speeds of passenger cars and the percentage of following vehicles observed in two-lane highways were applied to the model as the input. The revised relationships were developed from the computer simulation. RESULTS : In the revised average travel speed vs. flow relationship, the free-flow speed of 90km/h and 70km/h were added. It shows that the relationship between time delayed-rate and flow appeared to be appropriate with the log-function form and that there was no difference in time-delayed rate between the free flow speeds. In addition to revise the relationships, the speed prediction model and the time-delayed rate prediction model were also developed. CONCLUSIONS : The revised relationships between average travel speed, time-delayed rate, and flow would be useful in estimating the Level of Service(LOS) of a two-lane highway.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.4
• /
• pp.195-204
• /
• 2018

Lane keeping assist and departure prevention system are the key functions of ADAS. In this paper, we propose lane detection method which uses Gaussian function based RANSAC. The proposed method consists mainly of IPM (inverse perspective mapping), Canny edge detector, and Gaussian function based RANSAC (Random Sample Consensus). The RANSAC uses Gaussian function to extract the parameters of straight or curved lane. The proposed RANSAC is different from the conventional one, in the following two aspects. One is the selection of sample with different probability depending on the distance between sample and camera. Another is the inlier sample score that assigns higher weights to samples near to camera. Through simulations, we show that the proposed method can achieve good performance in various of environments.

• Journal of Drive and Control
• /
• v.17 no.1
• /
• pp.13-20
• /
• 2020

Adaptive neural networks based lateral controller is presented to guarantee path following performance for vehicle lane keeping in the presence of parameter time-varying characteristics of the vehicle lateral dynamics due to the road surface condition, load distribution, tire pressure and so on. The proposed adaptive controller could compensate vehicle lateral dynamics deviated from nominal dynamics resulting from parameter variations by incorporating it with neural networks that have the ability to approximate any given nonlinear function by adjusting weighting matrices. The controller is derived by using Lyapunov-based approach, which provides adaptive update rules for weighting matrices of neural networks. To show the superiority of the presented adaptive neural networks controller, the simulation results are given while comparing with backstepping controller chosen as the baseline controller. According to the simulation results, it is shown that the proposed controller can effectively keep the vehicle tracking the pre-given trajectory in high velocity and curvature with much accuracy under parameter variations.

• International Journal of Highway Engineering
• /
• v.17 no.5
• /
• pp.47-56
• /
• 2015

PURPOSES: The purpose of this study is to estimate the impact of variable message signage (VMS) on traffic safety as a function of road curve radius using statistical methods. METHODS: In order to analyze the impact of VMS installations on traffic safety, travel speed, lateral distance, and geometric data relating to road curvature in each study area was acquired and analyzed for the impact of providing VMS information on driver performance and traffic safety using statistical methods including student t-test, Mann-Whitney test, and the Anderson-Darling test for estimating traffic safety hazard zone in each lane. RESULTS: As a result of analyzing driver performance characteristics before and after providing VMS information, it was determined that by providing VMS information, mean travel speed is deceased and vehicles are driven with increased precision, following the centerline in the first and second lanes. Also the results of analyzing traffic safety impacts of VMS indicate that traffic safety performance factors in the first lane of the Gapyeong section can, on average, increase in the left and right side of the lane by 19.22% and 68.98%, respectively, and in the case of the second lane, safety impacts, on average, can increase in both sides by 100%. For the Hongcheon section, traffic safety impacts in the first lane, on average, can increase along the left and right sides of the lane by 32.31% and 47.18%, and within the second lane, traffic safety can be increased along the left and right side of the lane by 10.97% and -0.01%, respectively. CONCLUSIONS: Based on the results of this study, the impact on traffic safety obtained by providing VMS information for road sections with smaller curve radii is greater than can be obtained for road sections with larger curve radii.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.2
• /
• pp.81-86
• /
• 2015

GPS position errors were corrected for guidance of autonomous vehicles. From the vision, we can obtain the lateral distance from the center of lane and the angle difference between the left and right detected line. By using a controller which makes these two measurements zero, a lane following system can be easily implemented. However, the problem is that if there's no lane, such as crossroad, the guidance system of autonomous vehicle does not work. In addition, Line detection has problems working on curved areas. In this case, the lateral distance measurement has an error because of a modeling mismatch. For this reason, we propose GPS error correction filter based on curve-modeled lane detection and evaluated the performance applying it to an autonomous vehicle at the test site.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.21 no.2
• /
• pp.81-86
• /
• 2013

This paper presents the lane change system for collision avoidance. The proposed algorithm for the collision avoidance consists of path generation and path following. Using a calculated TTC (Time to Collision), partial braking is operated and collision avoidance path is generated considering relative distance, velocity and acceleration. Based on the collision avoidance path, desired yaw angle and yaw rate are calculated for the automated path following. The lateral controller is designed by a Lyapunov function approach using 3 D.O.F vehicle model and vehicle parameters. The required steering angle is determined from wheel velocity, longitudinal and lateral velocity in order to follow the desired yaw angle and yaw rate. This system is developed MATLAB/Simulink and its performance is evaluated using the commercial software CarSim.

• Journal of Korean Society of Transportation
• /
• v.6 no.1
• /
• pp.43-53
• /
• 1988

This study is to find the proper method of estimating urban road traffic capacity. To estimate road traffic capacity, the following methods are chosen ; 1) crossing point of Q-V and S-V, 2) critical velocity and density of Q-V-K model, 3) V-K model with density parameter. The density estimated through S-V relation is 174 veh./km. The methods used in this paper yields more stable values with 2286 veh./h/ in average. The estimated average capacity by three methods are 2272 veh./h. in multilane road. 2411 veh./h in three lane road and 2185 veh./h. in two lane road.