• Journal of Ocean Engineering and Technology
• /
• v.36 no.1
• /
• pp.41-49
• /
• 2022

With the advancement of intelligent vehicles and unmanned systems, there is a growing interest in underwater surveys using autonomous marine vehicles (AMVs). This study presents an automated planning strategy for a long-term survey mission using a fleet of AMVs consisting of autonomous surface vehicles and autonomous underwater vehicles. Due to the complex nature of the mission, the actions of the vehicle must be of high-level abstraction, which means that the actions indicate not only motion of the vehicle but also symbols and semantics, such as those corresponding to deploy, charge, and survey. For automated planning, the planning domain definition language (PDDL) was employed to construct a mission planner for realizing a powerful and flexible planning system. Despite being able to handle abstract actions, such high-level planners have difficulty in efficiently optimizing numerical objectives such as obtaining the shortest route given multiple destinations. To alleviate this issue, a widely known technique in operations research was additionally employed, which limited the solution space so that the high-level planner could devise efficient plans. For a comprehensive evaluation of the proposed method, various PDDL-based planners with different parameter settings were implemented, and their performances were compared through simulation. The simulation result shows that the proposed method outperformed the baseline solutions by yielding plans that completed the missions more quickly, thereby demonstrating the efficacy of the proposed methodology.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1223-1237
• /
• 2024

Autonomous vehicles use onboard sensors to sense the surrounding environment. In complex autonomous driving scenarios, the detection and recognition capabilities are constrained, which may result in serious accidents. An efficient way to enhance the detection and recognition capabilities is establishing collaborations with the neighbor vehicles. However, the collaborations introduce additional challenges in terms of the data heterogeneity, communication cost, and data privacy. In this paper, a novel personalized federated learning framework is proposed for addressing the challenges and enabling efficient collaborations in autonomous driving environment. For obtaining a global model, vehicles perform local training and transmit logits to a central unit instead of the entire model, and thus the communication cost is minimized, and the data privacy is protected. Then, the inference similarity is derived for capturing the characteristics of data heterogeneity. The vehicles are divided into clusters based on the inference similarity and a weighted aggregation is performed within a cluster. Finally, the vehicles download the corresponding aggregated global model and train a personalized model which is personalized for the cluster that has similar data distribution, so that accuracy is not affected by heterogeneous data. Experimental results demonstrate significant advantages of our proposed method in improving the efficiency of collaborative perception and reducing communication cost.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.11
• /
• pp.3523-3543
• /
• 2022

Autonomous vehicles are gradually being regarded as the mainstream trend of future development of the automobile industry. Autonomous driving networks generate many intensive and delay-sensitive computing tasks. The storage space, computing power, and battery capacity of autonomous vehicle terminals cannot meet the resource requirements of the tasks. In this paper, we focus on the task scheduling problem of autonomous driving in software-defined edge networks. By analyzing the intensive and delay-sensitive computing tasks of autonomous vehicles, we propose an emotion model that is related to task urgency and changes with execution time and propose an optimal base station (BS) task scheduling (OBSTS) algorithm. Task sentiment is an important factor that changes with the length of time that computing tasks with different urgency levels remain in the queue. The algorithm uses task sentiment as a performance indicator to measure task scheduling. Experimental results show that the OBSTS algorithm can more effectively meet the intensive and delay-sensitive requirements of vehicle terminals for network resources and improve user service experience.

• Convergence Security Journal
• /
• v.19 no.1
• /
• pp.19-30
• /
• 2019

As the 4th industrial revolution has recently become a hot topic, the importance of autonomous vehicles has increased and interest has been increasing worldwide, and accidents involving autonomous vehicles have also occurred. With the development of autonomous vehicles, the possibility of a cyber-hacking threat to the car network is increasing. Various countries, including the US, UK and Germany, have developed guidelines to counter cyber-hacking of autonomous vehicles, In the case of Korea, limited temporary operation of autonomous vehicles is being carried out, but the legal system to be applied in case of accidents caused by vehicle network hacking is insufficient. In this paper, based on the existing legal system, we examine the civil liability caused by the cyber hacking of the autonomous driving car, while we propose a law amendment suited to the characteristics of autonomous driving car and a legal system improvement plan that can give sustainable trust to autonomous driving car.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.3
• /
• pp.147-159
• /
• 2023

In order to make autonomous vehicles more trustworthy, it is necessary to focus on the users of autonomous vehicles. By evaluating the preferences for driving behaviors of autonomous vehicles, we aim to identify driving behaviors that increase the acceptance of users in autonomous vehicles. We implemented two driving behaviors, aggressive and cautious, in a driving simulator and allowed users to experience them. Biometric data was collected during the ride, and pre- and post-riding surveys were conducted. Subjects were categorized into two groups based on their driving habits and analyzed against the collected biometric data. Both aggressive and cautious driving subjects preferred the cautious driving behavior of autonomous vehicles.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.65-74
• /
• 2024

In recent years, image processing technology has played a critical role in the field of autonomous driving. Among them, image recognition technology is essential for the safety and performance of autonomous vehicles. Therefore, this paper aims to develop a hybrid image recognition system to enhance the safety and performance of autonomous vehicles. In this paper, various image recognition technologies are utilized to construct a system that recognizes and tracks objects in the vehicle's surroundings. Machine learning and deep learning algorithms are employed for this purpose, and objects are identified and classified in real-time through image processing and analysis. Furthermore, this study aims to fuse image processing technology with vehicle control systems to improve the safety and performance of autonomous vehicles. To achieve this, the identified object's information is transmitted to the vehicle control system to enable appropriate autonomous driving responses. The developed hybrid image recognition system in this paper is expected to significantly improve the safety and performance of autonomous vehicles. This is expected to accelerate the commercialization of autonomous vehicles.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.1
• /
• pp.63-73
• /
• 2024

Autonomous vehicles at Levels 3 to 5, currently under global research and development, seek to replace the driver's perception, judgment, and control processes with various sensors integrated into the vehicle. This integration enables artificial intelligence to autonomously perform the majority of driving tasks. However, autonomous vehicles currently obtain temporary driving permits, allowing them to operate on roads if they meet minimum criteria for autonomous judgment abilities set by individual countries. When autonomous vehicles become more widespread in the future, it is anticipated that buyers may not have high confidence in the ability of these vehicles to avoid hazardous situations due to the limitations of temporary driving permits. In this study, we propose a method for grading the judgment abilities of autonomous vehicles based on a driving simulator experiment comparing and evaluating drivers' abilities to avoid hazardous situations. The goal is to derive evaluation criteria that allow for grading based on specific scenarios and to propose a framework for grading autonomous vehicles. Thirty adults (25 males and 5 females) participated in the driving simulator experiment. The analysis of the experimental results involved K-means cluster analysis and independent sample t-tests, confirming the possibility of classifying the judgment abilities of autonomous vehicles and the statistical significance of such classifications. Enhancing confidence in the risk-avoidance capabilities of autonomous vehicles in future hazardous situations could be a significant contribution of this research.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.6
• /
• pp.224-239
• /
• 2018

Autonomous vehicles can significantly reduce accidents due to 'driver's carelessness', which occupies the majority of causes for traffic accidents, but they may fail to avoid traffic accidents due to unexpected situations, such as "trolley dilemma", vehicle defects and road defects. Therefore in situations Autonomous vehicles need to be made ethical choices. This study assumes that Autonomous vehicles can not avoid traffic accidents due to unexpected sink holes. In this situation, the traffic accident costs was analyzed for the ethical choices of Autonomous vehicles. In the process, Autonomous vehicles were made to choose one of three ethical choices : (1) Egoism with priority on passenger safety, (2) Deontology for minimizing human damages, (3) Utilitarianism with minimizing traffic accident costs. As a result of the analysis, egoism had the highest traffic accident costs, and deontology for minimizing human damages had the lowest traffic accident costs.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.1
• /
• pp.192-207
• /
• 2023

Commercialization of level-4 (Lv.4) autonomous driving applications requires the definition of a safe road environment under which autonomous vehicles can operate safely. Thus, a risk assessment model is required to determine whether the operation of autonomous vehicles can provide safety to is sufficiently prepared for future real-life traffic problems. Although the risk factors of autonomous vehicles were selected and graded, the decision-making method was applied as qualitative data using a survey of experts in the field of autonomous driving due to the cause of the accident and difficulty in obtaining autonomous driving data. The fuzzy linguistic representation of decision-makers and the fuzzy analytic hierarchy process (AHP), which converts uncertainty into quantitative figures, were implemented to compensate for the AHP shortcomings of the multi-standard decision-making technique. Through the process of deriving the weights of the upper and lower attributes, the road alignment, which is a physical infrastructure, was analyzed as the most important risk factor in the operation risk of autonomous vehicles. In addition, the operation risk of autonomous vehicles was derived through the example of the risk of operating autonomous vehicles for the 5 areas to be evaluated.

• Journal of Auto-vehicle Safety Association
• /
• v.16 no.2
• /
• pp.6-13
• /
• 2024

This paper presents a method of longitudinal motion planning of autonomous vehicles to deal with the non-signalized crosswalk environment. Based on the traffic laws, vehicles should slow down when passing the non-signalized crosswalk to prepare for situations where a nearby pedestrian starts to cross. If a pedestrian is in the crossing phase, vehicles should stop in front of the stop-line and wait until the pedestrian finishes the crossing maneuver. To realize these behaviors in autonomous vehicles, the driving mode and corresponding driving strategy are determined when vehicles encounter the crosswalk. The driving mode is determined according to the behavioral status of the nearby pedestrian. Longitudinal motion for the stopping or passing maneuver is planned according to the determined driving mode. The proposed algorithm has been validated via autonomous driving tests with our test vehicle in a real world. The test results show that the proposed algorithm enables the test vehicle to follow the traffic laws and behave safely against crossing pedestrians in the non-signalized crosswalk.