• Journal of Digital Convergence
• /
• v.19 no.10
• /
• pp.241-246
• /
• 2021

Although there have been many studies on laws and systems for the proliferation of autonomous vehicles, studies on the activation of autonomous vehicles from a strategic perspective are insufficient. This study examines the issues and problem solving methods of autonomous vehicles. Based on this, plans to activate autonomous vehicles from a strategic point of view are proposed. In order to solve the issues and problems of autonomous vehicles, it is necessary to clearly establish legal and institutional standards based on the reinforcement of the safety of autonomous vehicles. In the event of a traffic accident, who is responsible for the accident and responsibility for compensation should be prioritized. Diffusion strategies are established according to the level of autonomous driving for the activation of autonomous vehicles in strategic perspective. In addition, governmental support policies should be used as triggers for initial activation, and marketing mix strategies should be implemented based on segmentation, targeting, and positioning strategies.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.63-71
• /
• 2024

Deep reinforcement learning (RL) significantly influences autonomous vehicle development by optimizing decision-making and adaptation to complex driving environments through simulation-based training. In deep RL, an activation function is used, and various activation functions have been proposed, but their performance varies greatly depending on the application environment. Therefore, finding the optimal activation function according to the environment is important for effective learning. In this paper, we analyzed nine commonly used activation functions for RL to compare and evaluate which activation function is most effective when using deep RL for autonomous vehicles to learn highway merging. To do this, we built a performance evaluation environment and compared the average reward of each activation function. The results showed that the highest reward was achieved using Mish, and the lowest using SELU. The difference in reward between the two activation functions was 10.3%.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.75-81
• /
• 2022

Autonomous vehicles, which can dramatically solve the lack of parking spaces, are making great progress through deep reinforcement learning. Activation functions are used for deep reinforcement learning, and various activation functions have been proposed, but their performance deviations were large depending on the application environment. Therefore, finding the optimal activation function depending on the environment is important for effective learning. This paper analyzes 12 functions mainly used in reinforcement learning to compare and evaluate which activation function is most effective when autonomous vehicles use deep reinforcement learning to learn parking. To this end, a performance evaluation environment was established, and the average reward of each activation function was compared with the success rate, episode length, and vehicle speed. As a result, the highest reward was the case of using GELU, and the ELU was the lowest. The reward difference between the two activation functions was 35.2%.

• The Journal of Society for e-Business Studies
• /
• v.24 no.2
• /
• pp.91-112
• /
• 2019

In accordance with the advent of the 5th generation (5G) communication technology, we are having a change in various communication services which converge with high technologies related to the 4th Industrial Revolution. To utilize the upcoming 5G technology effectively and practically, we analyzed the technologies which have the most potential in convergence under the introduction of 5G technology and as a result, it is a autonomous vehicle that we'll discuss the core technologies of the 4th Industrial Revolution, which can lead to service activation by being combined with 5G technology. In addition, we developed an industry-service classification of 5G-based autonomous vehicle, we provided a basis for supporting a new business and its new business model converged with 5G communication technology. Furthermore, we will create a linkage matrix with the industry-service classification system of a new autonomous vehicles. This matrix will service as a guideline for industry-service development where autonomous vehicles can be utilized actively in the next generation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.42 no.4
• /
• pp.519-526
• /
• 2022

Basically, autonomous vehicles drive using road and traffic information collected by various sensors. However, it is known that there is a limitation to realizing fully autonomous driving with only such technologies and information. In recent, various efforts are being made to overcome the limitations of sensor-based autonomous driving, and efforts are also underway to utilize more specific and accurate road and traffic information, called local dynamic map (LDM). However, LDM-related data standards and specifications have not yet been sufficiently verified, and research on the spatiotemporal scope of LDM during autonomous driving is extremely limited. Based on this background, the purpose of this study is to identify these limitations through an analysis of previous LDM-related studies and to present a framework for calculating the spatiotemporal activation section of LDM-based road and traffic information.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.23 no.2
• /
• pp.173-188
• /
• 2024

With the advancement of autonomous vehicle (AV) technology, autonomous driving on real roads has become feasible. However, there are challenges in achieving complete autonomy due to perceptual blind areas, which occur when the AV's sensory range or capabilities are limited or impaired by surrounding objects or environmental factors. This study aims to analyze AV accident patterns and safety issues of perceptual blind area that may occur in urban areas, with the goal of developing test scenarios for Level 4+ autonomous driving. It utilized AV accident data from the California Department of Motor Vehicles (DMV) to compare accident patterns and characteristics between AVs and conventional vehicles based on activation status of autonomous mode. It also categorized AV disengagement data to identify types and real-world cases of disengagements caused by perceptual blind areas. The analysis revealed that AVs exhibit different accident types due to their safe driving maneuvers, and three types of perceptual blind area scenarios were identified. The findings of this study serve as crucial foundational data for developing Level 4+ autonomous driving test scenarios, enabling the design of efficient strategies to mitigate perceptual blind areas in various scenarios. This, in turn, is expected to contribute to the effective evaluation and enhancement of AV driving safety on real roads.

• Journal of Korean Library and Information Science Society
• /
• v.52 no.1
• /
• pp.27-54
• /
• 2021

This study examines the overall contents of robots, drones, and autonomous driving that can be applied to libraries among autonomous objects, and proposes a plan that can be introduced and applied to libraries in the future based on this. As a result of the study, in the case of the building, robots and drones can be used to apply from collection inspection, collection transport, collection arrangement, collection classification, book location guidance, book recommendation, loan/return, library general guidance, and reference information service. Outside of the building, robots, drones, and autonomous vehicles can be used for book delivery service, book return service, and unmanned mobile libraries. This study is a basic research for the introduction and application of autonomous objects in the library, and follow-up studies such as perception survey and application model development for systematic introduction should be conducted in the future.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.1
• /
• pp.291-310
• /
• 2023

This study explores the factors which affect user acceptance for Robo Taxi, an electricity-based Autonomous Vehicles based on a Value based Adoption Model. The three main factors of benefit (usefulness and enjoyment), sacrifice (technicality and perceived fee level), and user experience about mobility services such as car sharing, taxi, and autonomous vehicles, were finally selected as independent variables as a influential factors on perceived values and adoption intention of Robo taxi. The study found that usefulness, enjoyment, and perceived fee had a significant effects on adoption intention, and some user experiences had a significant effect on benefit factors. This study has important implications for incorporating the Value-based Adoption Model results into the service design for the activation of Robo taxi, and furthermore, they can provide a theoretical basis for effective use of the research findings.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.101-111
• /
• 2022

Conditional autonomous vehicles should hand over control to the driver according on driving situations. However, if the driver is immersed in a non-driving task, the driver may not be able to make suitable decisions. Previous studies have confirmed that the cues enhance take-over performance with a directional information on driving. However, studies on the effect of take-over cues on the driver's brain activities are rigorously investigated yet. Therefore, this study we evaluates the driver's brain activity according to the take-over cue. A total of 25 participants evaluated the take-over performance using a driving simulator. Brain activity was evaluated by functional near-infrared spectroscopy, which measures brain activity through changes in oxidized hemoglobin concentration in the blood. It evaluates the activation of the prefrontal cortex (PFC) in the brain region. As a result, it was confirmed that the driver's PFC was activated in the presence of the cue so that the driver could stably control the vehicle. Since this study results confirmed that the effect of the cue on the driver's brain activity, and it is expected to contribute to the study of take-over performance on biomakers in conditional autonomous driving in future.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.1
• /
• pp.1-9
• /
• 2019

In this paper, we propose a method of generating a 2D gray image from LiDAR 3D reflection intensity. The proposed method uses the Fully Convolutional Network (FCN) to generate the gray image from 2D reflection intensity which is projected from LiDAR 3D intensity. Both encoder and decoder of FCN are configured with several convolution blocks in the symmetric fashion. Each convolution block consists of a convolution layer with $3{\times}3$ filter, batch normalization layer and activation function. The performance of the proposed method architecture is empirically evaluated by varying depths of convolution blocks. The well-known KITTI data set for various scenarios is used for training and performance evaluation. The simulation results show that the proposed method produces the improvements of 8.56 dB in peak signal-to-noise ratio and 0.33 in structural similarity index measure compared with conventional interpolation methods such as inverse distance weighted and nearest neighbor. The proposed method can be possibly used as an assistance tool in the night-time driving system for autonomous vehicles.