• The Journal of Korea Robotics Society
• /
• v.12 no.3
• /
• pp.297-305
• /
• 2017

As the development of autonomous vehicles becomes realistic, many automobile manufacturers and components producers aim to develop 'completely autonomous driving'. ADAS (Advanced Driver Assistance Systems) which has been applied in automobile recently, supports the driver in controlling lane maintenance, speed and direction in a single lane based on limited road environment. Although technologies of obstacles avoidance on the obstacle environment have been developed, they concentrates on simple obstacle avoidances, not considering the control of the actual vehicle in the real situation which makes drivers feel unsafe from the sudden change of the wheel and the speed of the vehicle. In order to develop the 'completely autonomous driving' automobile which perceives the surrounding environment by itself and operates, ability of the vehicle should be enhanced in a way human driver does. In this sense, this paper intends to establish a strategy with which autonomous vehicles behave human-friendly based on vehicle dynamics through the reinforcement learning that is based on Q-learning, a type of machine learning. The obstacle avoidance reinforcement learning proceeded in 5 simulations. The reward rule has been set in the experiment so that the car can learn by itself with recurring events, allowing the experiment to have the similar environment to the one when humans drive. Driving Simulator has been used to verify results of the reinforcement learning. The ultimate goal of this study is to enable autonomous vehicles avoid obstacles in a human-friendly way when obstacles appear in their sight, using controlling methods that have previously been learned in various conditions through the reinforcement learning.

• Journal of Korean Society of Transportation
• /
• v.17 no.4
• /
• pp.71-84
• /
• 1999

The purpose of this study is to construct a dynamic traffic analysis model using the existing traffic flow theory in order to develope a dynamic traffic assignment technique. In this study the dynamic traffic analysis model was constructed using Daganzo's CELL TRANSMISSION THEORY which was considered more suitable to dynamic traffic assignment than the other traffic flow theories. We developed newly the diverging split module, the cost update module and the link cost function and defined the maximum waiting time decision function that Daganzo haven't defined certainly at his Papers. The output that resulted from the simulation of the dynamic traffic analysis model with test network I and II was shown at some tables and figures, and the analysis of the bottleneck and the HOV lane theory showed realistic outputs. Especially, the result of traffic assignment using the model doesn't show equilibrium status every time slice but showed that the average travel cost of every path maintains similarly in every time slice. It is considered that this model can be used at the highway operation and the analysis of traffic characteristics at a diverging section and the analysis of the HOV lane effect.

• Journal of Korean Society of Transportation
• /
• v.20 no.1
• /
• pp.131-139
• /
• 2002

This study introduces a developed artificial neural networks(ANN) model as a more efficient and reliable prediction model in operating speed Prediction with the 85th percentile horizontal curve of two-way rural highway in the aspect of evaluating highway design consistency. On the assumption that the speed is decided by highway geometry features, total 30 survey sites were selected. Data include currie radius, curve length, intersection angle, sight distance, lane width, and lane of those sites and were used as input layer data of the ANN. The optimized model structure was drawn by number of unit of hidden layer, learning coefficient, momentum coefficient, and change in learning frequency in multi-layer a ANN model. To verify learning Performance of ANN, 30 survey sites were selected while data in obtained from the 20 cites were used as learning data and those from the remaining 10 sites were used as predictive data. As a result of statistical verification, the model D of 4 types of ANN was evaluated as the most similar model to the actual operating speed value: R2 was 85% and %RMSE was 0.0204.

• Journal of Korean Society of Transportation
• /
• v.20 no.1
• /
• pp.121-130
• /
• 2002

Purpose of this paper is proposing technical evaluation method for safety on two-lane rural highways, this method can contribute to an establishment of the safety evaluation criteria and eventually contribute to the highway safety. In this study, accident data and field data(grade, radius, speed factor) are collected and are analyzed. And new technical evaluation method is proposed by problem in review of existent methods and analysis result. Property of new method is verified by comparing with existent one. Following are results of this study. 1) Proposed technical evaluation method is Proposed by using difference of speed variance and difference of acceleration. 2) In range of field survey, proposed method is more proper than existent one by comparing. 3) Headways under 6 second affect speed of next vehicle. 4) A point with the lowest speed in curve is placed about 35m from the beginning point. 5) Correlation of speed variance and severity of highway(EPDO) is statistically significant. For using proposed method, more tests are needed on other locations with various geometries. Moreover, advanced study is needed about classifying and giving the weight of each criterion.

• International Journal of Highway Engineering
• /
• v.11 no.2
• /
• pp.217-228
• /
• 2009

Sections under the influence of merging in an uninterrupted facility create irregular interaction between vehicles, such as lane change, speed acceleration and deceleration because of the merging of ramp traffic flows which have traffic characteristics different from those of the main line. This causes a confused traffic flow phenomenon(turbulence), which is considered an unstable traffic characteristic between various continuous points in consideration of v conditions. In this study, in merge influence sections, detectors by lane-point were installed to create time and space-series -traffic data. The least significant difference(LSD), as the criteria for discriminating a significant speed change between points, was calculated to examine the turbulence. As a result, turbulence in merge influence section was found to change the zones of such occurrence and the seriousness levels according to traffic condition. Thus, the maximum merge influence section due to the turbulence was created in the traffic condition before congestion when traffic increases. According to characteristics of changes in speed, merge influence section was divided into upstream 100m$\sim$downstream 100m(a section of speed reduction), and downstream 100m$\sim$downstream 400m(a section of reduced speed maintenance and acceleration).

• International Journal of Highway Engineering
• /
• v.10 no.2
• /
• pp.57-67
• /
• 2008

At present engineers use the highway design speed concept to develop the horizontal alignment. This concept has a strength of attaining consistent horizontal alignments because of its use of a single speed value. Yet it shows a critical weakness that when opened the operating speed for the vehicles on the road can be a lot different than the design speed. To resolve this, many countries already develop the horizontal alignment by adopting procedures that weigh vehicle operating speeds, and this research joins them by developing an operating speed based horizontal alignment design. We have collected vehicle speed samples over the nation, selected some speed influential design elements by doing a statistical analysis, provided a set of models for two-lane roads and four lane roads, and showed a stepwise feedback procedure by doing a case study. It is underscored that in the case study the proposed procedure has scaled down the speed inconsistency problem, and we are of opinion that our procedure would coin both investment efficiency and speed consistency in future highway projects.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.23 no.6 s.165
• /
• pp.931-942
• /
• 1999

This paper develops a 3 DOF vehicle model which includes lateral, roll and yaw motion to study a 4WS vehicle. The model is used for the simulation of a 4WS vehicle behavior, and to derive a control algorithm for rear wheel steering. This paper uses a feedforward plus feedback control scheme to compute a rear wheel steering angle. The feedforward control scheme for computing the first rear wheel steering angle uses a gain which is acquired by multiplying a proper value on a gain to maintain a zero sideslip angle. The feedback control scheme for computing the second rear wheel steering angle uses fuzzy logic and model following control scheme. A linear 2 DOF model is used as a reference model for model following control, and is derived from the developed 3 DOF model by neglecting sprung mass roll motion. A reference state variable is yaw rate, and is computed using the linear 2 DOF model. J-turn and lane change maneuver simulation are performed to show the effectiveness of the developed control scheme. The simulation results show that the 4WS vehicle with the developed control scheme has much better performance in yaw rate, lateral acceleration, roll angle, and sideslip angle than the 2WS vehicle. Also, the results show that the performance of the developed control is close to the one of an optimal control which assumes all states are perfect.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.21 no.4
• /
• pp.683-690
• /
• 2012

In this work, the semi-active control of a large-sized bus suspension system with an MR damper was studied. An MR damper model that can aptly describe the hysteretic characteristics of an MR damper was adopted. Parameter values of the MR damper model were suitably modified by considering the maximum damping force of a passive damper used in the suspension system of a real large-sized bus. In addition, a fuzzy logic controller was developed for semi-active control of a suspension system with an MR damper. The vertical acceleration at the attachment point of the MR damper and the relative velocity between sprung and unsprung masses were used as input variables, while voltage was used as the output variable. Straight-ahead driving simulations were performed on a road with a random road profile and on a flat road with a bump. In straight-ahead driving simulations, the vertical acceleration and pitch angle were measured to compare the riding performance of a suspension system with a passive damper with that of a suspension with an MR damper. In addition, a single lane change simulation was performed. In the simulation, the lateral acceleration and roll angle were measured in order to compare the handling performance of a suspension system using a passive damper with that of a suspension system using an MR damper.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.9 no.4
• /
• pp.45-51
• /
• 2009

The paper presents the economic analysis of optimized-rehabilitation area method considered as one of the preventive maintenance methods in asphalt concrete pavement. The optimized-rehabilitation area was selected based on the analysis of traffic lane characteristics. The main concept of the selected method was to minimize the maintenance cost. The effective width of traffic lane in this method was 70 cm of each wheel path. According to the traffic survey conducted in this research, more than 95% of vehicles passed within the width of each wheel path. The new preventive optimized-rehabilitation area method showed less maintenance cost than the conventional overlay. In addition, traffic congestions and the user cost can be reduced. The research results revealed that the total maintenance cost was reduced by 35% by using the new method compare to the conventional one.

• Journal of Korean Society of Transportation
• /
• v.17 no.2
• /
• pp.149-162
• /
• 1999

One of the most important steps of the design, capacity and operation analysis stapes in the two-lane two way highways is the effect of heavy vehicle to traffic flow quality. This heavy vehicle's effect on traffic flow can be represented as PCE, which is the number of passenger cars that are displaced by a single heavy vehicle of a particular type under prevailing roadway, traffic, and control conditions. In this paper, we focus on the heavy vehicles effect on volume, speed, delay, and the maneuver of freedom which are major MOE's in traffic operation analysis and PCE criterion which should be measurable, determinable and able to reflect the traffic flow characteristics. Therefore, the objective of the paper is to determine the PCE criterion and to develop a new PCE determination method. In this study, delay is adopted as PCE criterion and, for calculation of delay, the highway is divided into the passing zone and the no-passing zone. PCE is determined by comparing the delay due to total traffic flow interaction with the delay due to a single heavy vehicle, Also, this paper proposes a new method to determine the average PCE on the highway that has the passing zones and no-passing zones.