• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.316-322
• /
• 2023

The increasing popularity of autonomous unmanned aerial vehicles (UAVs) can be attributed to their wide range of applications. 3D path planning is one of the crucial components enabling autonomous flight. In this paper, we present a novel 3D path planning algorithm that generates and utilizes curvature-based trajectories. Our approach leverages circular properties, offering notable advantages. First, circular trajectories make collision detection easier. Second, the planning procedure is streamlined by eliminating the need for the spline process to generate dynamically feasible trajectories. To validate our proposed algorithm, we conducted simulations in Gazebo Simulator. Within the simulation, we placed various obstacles such as pillars, nets, trees, and walls. The results demonstrate the efficacy and potential of our proposed algorithm in facilitating efficient and reliable 3D path planning for UAVs.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.3
• /
• pp.387-392
• /
• 2014

Now a days, there are various application functions to transmit from vehicles to the Internet and vice versa. And the communication can be operated through a roadside infrastructure including with possible use of routing protocols. Specifically, autonomous vehicles for remote driving and monitoring requires transmitting of high depth of multimedia such as video. Especially in a populated urban area, an efficient network is vital because of handling a great amount of the data. Therefore, in this paper, efficient network topology for a crossroad in urban area is suggested by performance evaluation of vehicular networks using a wireless LAN and a routing protocol. For the performance evaluation, various vehicular network topologies are designed and simulated in OPNet simulator.

• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.147-159
• /
• 2020

Over the last years, a number of different path following methods for the autonomous parking system have been proposed for tracking planned paths. However, it is difficult to find a study comparing path following methods for a short path length with large curvature such as a parking path. In this paper, we conduct a comparative study of the path following methods for perpendicular parking. By using Monte-Carlo simulation, we determine the optimal parameters of each controller and analyze the performance of the path following. In addition, we consider the path following error occurred at the switching point where forward and reverse paths are switched. To address this error, we conduct the comparative study of the path following methods with the one thousand switching points generated by the Monte-Carlo method. The performance of each controller is analyzed using the V-rep simulator. With the simulation results, this paper provides a deep discussion about the effectiveness and limitations of each algorithm.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.1
• /
• pp.38-44
• /
• 2021

This study aimed to verify the criteria of the driver monitoring systems proposed by UNECE ACSF informal working group and the ministry of land, infrastructure, and transport of South Korea using driving behavior data. In order to verify the criteria, we investigated the safety regulations of driver monitoring systems in a conditional autonomous vehicle and found that the driver monitoring measures were related to eye blinks times, head movements, and eye closed duration. Thus, we took two different experimental data including real-world driving and simulator-based drowsy driving behaviors in previous studies. The real-world driving data were used for analyzing blink times and head movement intervals, and the drowsiness data were used for eye closed duration. In the real-world driving study, 52 drivers drove approximately 11.0 km of rural road (about 20 min), 7.9 km of urban road (about 25 min), and 20.8 km of highway (about 20 min). The results suggested that the appropriate number of blinks during the last 60 seconds was 4 times, and the head movement interval was 35 seconds. The results from drowsy driving data will be presented in another paper - part 2.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.1
• /
• pp.45-50
• /
• 2021

This study aimed to verify the criteria of the driver monitoring systems proposed by UNECE ACSF informal working group and the ministry of land, infrastructure, and transport of South Korea using driving behavior data. In order to verify the criteria, we investigated the safety regulations of driver monitoring systems in a conditional autonomous vehicle and found that the driver monitoring measures were related to eye blinks times, head movements, and eye closed duration. Thus, we took two different experimental data including real-world driving and simulator-based drowsy driving behaviors in previous studies. The real-world driving data were used for analyzing blink times and head movement intervals, and the drowsiness data were used for eye closed duration. In the drowsy driving study, 10 drivers drove approximately 37 km of a monotonous highway (about 22 min) twice. The results suggested that the appropriate duration of eyes continuously closed was 4 seconds. The results from real-world driving data were presented in the other paper - part 1.

• 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.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.641-652
• /
• 2021

Drivers have willing to do secondary tasks in situations deemed safe autonomous driving. I studied that positive notifications for secure areas could improve situation awareness and driving performance after TOR(Take over request) in autonomous driving. Comparing TOR alert only and monitoring alert conditions, participants in the positive notification condition showed higher situation awareness and driving performance. Also, in emotional assessment, the positive notification condition showed higher positive evaluation than other conditions. Due to Covid-19, I designed experiments separate online with driving videos in experiment 1 and offline using a driving simulator in experiment 2. This study has implications that presented a different perspective on autonomous driving notification design.

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.386-393
• /
• 2023

Recently, the demand for autonomous navigation technology has increased, and related research is also increasing. Autonomous ships generally follow the planned route, calculate the avoidance route according to the risk situation while sailing, and follow a calculated route. In general, an automatic steering device is used to follow the route, and among the operational automatic steering device methods, the route control mode is the most appropriate method to apply to autonomous ships. Therefore, in this study, we developed a route-tracking algorithm to apply an avoidance route using the navigation control mode of an automatic steering device. The algorithm was developed by dividing the straight and turning sections. A performance test was conducted to satisfy the performance suggested by IEC 62065, the relevant international standard, using simulator equipment that had acquired international certification to verify its performance. The results of the performance verification confirmed that the cross-track error, which represents the straight distance between the ship and the route, satisfied the performance standards suggested by IEC 62065 when the ship followed the route.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.109-123
• /
• 2024

Research into vision-based end-to-end autonomous driving systems utilizing deep learning and reinforcement learning has been steadily increasing. These systems typically encode continuous and high-dimensional vehicle states, such as location, velocity, orientation, and sensor data, into latent features, which are then decoded into a vehicular control policy. The complexity of urban driving environments necessitates the use of state representation learning through networks like Variational Autoencoders (VAEs) or Convolutional Neural Networks (CNNs). This paper analyzes the impact of different image state encoding methods on reinforcement learning performance in autonomous driving. Experiments were conducted in the CARLA simulator using RGB images and semantically segmented images captured by the vehicle's front camera. These images were encoded using VAE and Vision Transformer (ViT) networks. The study examines how these networks influence the agents' learning outcomes and experimentally demonstrates the role of each state representation technique in enhancing the learning efficiency and decision- making capabilities of autonomous driving systems.

• Journal of IKEEE
• /
• v.7 no.2 s.13
• /
• pp.127-134
• /
• 2003

Ad Hoc Networks, autonomous distributed network using routing scheme, does not operate properly owing to multi flow service when network load increases at specific network node. In this paper, we suggest traffic management routing protocol in Ad Hoc Network to reduce network traffic congestion and distribute network load in data transmission. Through test results of proposed algorithm under NS(Network Simulator)simulator environments . we acquired reduced network load and increased data transmission rate.