• International Journal of Highway Engineering
• /
• v.14 no.4
• /
• pp.149-162
• /
• 2012

PURPOSES : The causes of traffic accidents can be classified into the factors of highway users, vehicles, and driving environments. Traffic accidents result from the deficiency in single or combination of these three factors. The objective of this study is to define the "potentially hazardous sections of highway" in terms of traffic safety considering these three factors. METHODS : The test drivers performed repeated driving on these highway sections. The drivers and passengers recorded the sections on which the driving was uncomfortable, and the speeds on the sections excluding the uncomfortable sections were used for the development of the model. RESULTS : The model is composed of three sub-models for each of the horizontal curve, tangent, and the section where the curve starts/ends. The safe driving behavior coefficients by the horizontal curvature were derived by comparing the maximum operating speeds at which the vehicle may slide or deviate and the speeds at which the drivers feel comfort. The safety speeds on tangent were derived by the length of tangent section considering the driver's desired speeds under the traffic condition on which the drivers hardly influenced by the other vehicles. For the sections where the curve starts/ends, the driving behaviors were classified by the distances between the curves, and the safe acceleration/deceleration speeds were derived on which the drivers enter/exit the curve sections safely. CONCLUSIONS : Safety speed could then be regarded that the model suggested in this study may be useful to define the potentially hazardous highway section and contribute the improvement of highway safety.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.11
• /
• pp.76-88
• /
• 1998

While most of the research on reinforcement learning assumed a discrete control space, many of the real world control problems need to have continuous output. This can be achieved by using continuous mapping functions for the value and action functions of the reinforcement learning architecture. Two questions arise here however. One is what sort of function representation to use and the other is how to determine the amount of noise for search in action space. The ubiquitous neural network is used here to learn the value and policy functions. Next, the reinforcement predictor that is intended to predict the next reinforcement is introduced that also determines the amount of noise to add to the controller output. The proposed reinforcement learning architecture is found to have a sound on-line learning control performance especially at high-speed road following of high curvature road. Both computer simulation and actual experiments on a test vehicle have been performed and their efficiency and effectiveness has been verified.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.5
• /
• pp.435-444
• /
• 2005

This paper presents a lane-departure identification (LDI) algorithm for a traveling vehicle on a structured road. The algorithm makes up for the weak points of the former method based on EDF[1] by introducing a Lane Boundary Pixel Extractor (LBPE), the well known Hough transform, and liner regression. As a filter to extract pixels expected to be on lane boundaries, the LBPE plays an important role in enhancing the robustness of LDI. Utilizing the pixels from the LBPE the Hough transform provides the lane-related parameters composed of orientation and distance, which are used in the LDI. The proposed LDI is based on the fact the lane-related parameters of left and right lane boundaries are symmetrical as for as the optical axis of a camera mounted on a vehicle is coincident with the center of lane; as the axis deviates from the center of lane, the symmetrical property is correspondingly lessened. In addition, the LDI exploits a linear regression of the lane-related parameters of a series of successive images. It plays the key role of determining the trend of a vehicle's traveling direction and minimizing the noise effect. Except for the two lane-related parameters, the proposed algorithm does not use other information such as lane width, a curvature, time to lane crossing, and of feet between the center of a lane and the optical axis of a camera. The system performed successfully under various degrees of illumination and on various road types.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.147-153
• /
• 2020

Conventional lane detection algorithms have problems in that the detection rate is lowered in road environments having a large change in curvature and illumination. The probabilistic Hough transform method has low lane detection rate since it exploits edges and restrictive angles. On the other hand, the method using a sliding window can detect a curved lane as the lane is detected by dividing the image into windows. However, the detection rate of this method is affected by road slopes because it uses affine transformation. In order to detect lanes robustly and avoid obstacles, we propose driving assist system using semantic segmentation based on deep learning. The architecture for segmentation is SegNet based on VGG-16. The semantic image segmentation feature can be used to calculate safety space and predict collisions so that we control a vehicle using adaptive-MPC to avoid objects and keep lanes. Simulation results with CARLA show that the proposed algorithm detects lanes robustly and avoids unknown obstacles in front of vehicle.

• International Journal of Automotive Technology
• /
• v.5 no.2
• /
• pp.77-88
• /
• 2004

Narrow commuter vehicles can address many congestion, parking and pollution issues associated with urban transportation. In making narrow vehicles safe, comfortable and acceptable to the public, active tilt control systems are likely to playa crucial role. This paper focuses on the development of an active direct tilt control system for a narrow vehicle that utilizes an actuator in the vehicle suspension. A simple PD controller can stabilize the tilt dynamics of the vehicle to any desired tilt angle. However, the challenges in the tilt control system design arise in determining the desired lean angle in real-time and in minimizing tilt actuator torque requirements. Minimizing torque requirements requires the tilting and turning of the vehicle to be synchronized as closely as possible. This paper explores two different control design approaches to meet these challenges. A Receding Horizon Controller (RHC) is first developed so as to systematically incorporate preview on road curvature and synchronize tilting with driver initiated turning. Second, a nonlinear control system that utilizes feedback linearization is developed and found to be effective in reducing torque. A close analysis of the complex feedback linearization controller provides insight into which terms are important for reducing actuator effort. This is used to reduce controller complexity and obtain a simple nonlinear controller that provides good performance.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1791-1796
• /
• 2005

Semi-shield tunneling is one of the propulsion construction methods used to lay pipes underground between two pits named 'entrance' and 'destination', respectively. Usually a simple composition, such as 'a fiducial target at the entrance+a total station (TS)+a target on the machine', is used to confirm the planned course. However, unavoidable curved sections are present in small-sized pipe lines, which are laid after implementation of a road system, for public works such as waterworks, sewer, electrical power, and gas and communication networks. Therefore, if the planned course has a curved section, it is difficult to survey the course with the abovementioned simple composition. This difficulty could be solved by using the multiple total stations (MTS), which attaches the cross type linear LED target to oneself. The MTS are disposed to where each TS can detect the LED target at the other TS or the base point or the machine. And the accurate relative positions between each MTS and target are calculated from measured data. This research proposes the relative and absolute coordinate calculation algorithm by using three MTS to measure a curved course with 20m curvature at 30m maximum distance, and verifies the algorithm experimentally.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.1
• /
• pp.93-103
• /
• 1999

To detect obstacles and lane-markers for driving vehicles, a new moving window scheme where moving windows are assigned to an image frame captured by a camera is addressed. For the detection of obstacles, it is important to estimate lane-markers precisely and rapidly. For this purpose, selecting some partes of an image frame at the expected lane locations, i.e., selecting window are generally adopted for extracting lane-markers efficiently. In this paper, a new scheme that extracts lane-markers precisely by assigning variable size windows at the expected locations of lane-markers considering the road curvature and finally detects obstacles within a driving lane is proposed. The accuracy improvement using this moving window scheme is showed by comparing to the conventional fixed window method and to using radar to laser sensors.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.26 no.4
• /
• pp.6-13
• /
• 2012

In this paper, we took an indirect reflectance type that irradiates light from an LED module, which reflects light from a reflector onto the road. This LED pedestrian lighting design produces minimal glare from a relatively simple structure. Using LightTools 7.2, we conducted a study on the varying shape and curvature of the reflector. In order to further investigate this data, we researched the ratio of the DLOR and ULOR, the average illuminance, and the uniformity of the illumination. Based on this research, we selected optimum conditions to realize the actual LED pedestrian lighting using RELUX software. We found the average illuminance value to be 22[1x] and the illuminance uniformity ratio to be 0.23, which met the criteria of the proposed standards set in KS C 7658. In addition, it also satisfied the criteria of the standards set in KS A 3701 by having TI (Threshold Increment) at 7[%].

• Steel and Composite Structures
• /
• v.27 no.2
• /
• pp.185-192
• /
• 2018

Fly ash can significantly improve concrete workability and performance, and recycling fly ash in concrete can contribute to a cleaner environment. Since fly ash influences pozzolanic reactions in concrete, mechanical behaviors of concrete containing fly ash differ from traditional concrete. Creep behaviors of fly ash concrete filled steel tube arch were experimentally investigated for 10% and 30% fly ash replacement. The axes of two arches are designed as circular arc with 2.1 m computed span, 0.24 m arch rise, and their cross-sections are all in circular section. Time dependent deflection and strain of loading and mid-span steel tube were measured, and long term deflection of the model arch with 10% fly ash replacement was significantly larger than with 30% replacement. Considering the steel tube strain, compressive zone height, cross section curvature, and internal force borne by the steel tube, the compressive zone height and structural internal forces increased gradually over time due to concrete creep. Increased fly ash content resulted in more significant neutral axis shift. Mechanisms for internal force effects on neutral axis height were analyzed and verified experimentally.

• Journal of Korean Society of Transportation
• /
• v.26 no.3
• /
• pp.187-197
• /
• 2008

While driving a highway, A driver gets lots of information through geometrical structure, traffic situation, signs on the road. He gets most of the information by visual sense. Acceleration or deceleration and driving direction depend on sight distance. Therefore, it's essential to secure a driver's sight distance for a safe drive. However, design guides of geometrical structure and sight distance suggest respective standards of horizontal and vertical alignment. They do not indicate quantitative standard of combined alignment. Currently, element separated on a two-dimensional projected plane are available, but they do not guarantee safe and pleasant design. I will use the existent model analysing three-dimensional sight distance through mathematical calculation and sort a variety of geometrical structure element and type. In these researches, we will look at how much three-dimensional sight distance is overestimated or underestimated compared to two-dimensional. I will develop a program which predicts traffic velocity on the curvature of two-lane provincial road. stopped sight distance and three-dimensional sight distance will be compared at a predicted drive velocity. I will suggest the way to evaluate road consistency.