• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2019.11a
• /
• pp.265-265
• /
• 2019

Marine accidents continue to occur every year due to human errors. The purpose of this study is to promote navigational safety by preventing ship collision accidents caused by human errors of behavior of navigators. There are two ways to manage human error caused by navigator's behavior. It is divided in individual approach and system approach, which is applied to situational awareness theory and Rasmussen's behavioral theory. This study investigated past marine accidents caused by human error and conducted experiments using ship handling simulators to identify these two behavioral characteristics. After analyzing two human error characteristics, we will propose a countermeasure in next study.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.163-168
• /
• 2002

The voting algorithm for action selection performs self-improvement by Reinforcement learning algorithm in the dynamic environment. The proposed voting algorithm improves the navigation of the robot by adapting the eligibility of the behaviors and determining the Command Set Generator (CGS). The Navigator that using a proposed voting algorithm corresponds to the CGS for giving the weight values and taking the reward values. It is necessary to decide which Command Set control the mobile robot at given time and to select among the candidate actions. The Command Set was learnt online by means as Q-learning. Action Selector compares Q-values of Navigator with Heterogeneous behaviors. Finally, real-world experimentation was carried out. Results show the good performance for the selection on command set as well as the convergence of Q-value.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.25 no.3
• /
• pp.320-327
• /
• 2019

The purpose of this paper is to compare seafarers' behavior according to traffic conditions of a road and an onshore locations. Behaviors are classified into three categories: Skill-, Rule- and knowledge-based mode. Experimental data were collected using the questionnaires for navigators, working in a merchant ship. To compare the behaviors, we used the four analysis method; the degree of frequency, reliability test, correlation and linear regression. As a result of the study, it was found that Skill-based behavior shows more higher in the road traffic than the maritime traffic, and rule-based behavior shows more higher in the maritime traffic than the road traffic. Also, the behavior in the navigation situation showed statistical significance. Especially, in the case of Rule-based behavior, a high correlation between road and maritime was found. This study can be expected to apply to complementary system utilization between error management system of onshore and maritime traffic.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.6
• /
• pp.701-708
• /
• 2018

Human error is an important cause of maritime accidents and the identification of human error is fundamental to maritime-accident preventions. In particular, the pattern of technical behavior taken in the circumstance of bridge teams(navigator & helmsman) provides important information to identify human error. The purpose of this study is to identify and analyze technical behavior pattern of bridge teams using Williamson's turn for rescue of persons overboard. The focus of this study is to build and analyze a cognitive model of the human behavior factors of the bridge teams in the process of implementing the experiments. The experimental environment was constructed using a ship-handling simulator and conducted an experiment on participants from 24 bridge teams. As a result of the experiment, it was able to identify the behavior pattern of the ship's maneuvering and maintain trajectory using the rudder and engine. This study is expected to correct human error in the bridge teams application to the certification and training of seafarers.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.39 no.1
• /
• pp.77-83
• /
• 2003

Autonomous mobile robots based on the Web have been already used in public places such as museums. There are many kinds of problems to be solved because of the limitation of Web and the dynamically changing environment. We present a methodology for intelligent mobile robot that demonstrates a certain degree of autonomy in navigation applications. In this paper, we focus on a mobile robot navigator equipped with neuro-fuzzy controller which perceives the environment, make decisions, and take actions. The neuro-fuzzy controller equipped with collision avoidance behavior and target trace behavior enables the mobile robot to navigate in dynamic environment from the start location to goal location. Most telerobotics system workable on the Web have used standard Internet techniques such as HTTP, CGI and Scripting languages. However, for mobile robot navigations, these tools have significant limitations. In our study, C# and ASP.NET are used for both the client and the server side programs because of their interactivity and quick responsibility. Two kinds of simulations are performed to verify our proposed method. Our approach is verified through computer simulations of collision avoidance and target trace.