Kwang-Ho Seok

A new robot motion authoring method using HTM

This paper presents a user-friendly authoring method for humanoid robots using HTM (hierarchical temporal memory), a machine-learning model. The proposed method is an intuitive scheme which allows users (especially children and the elderly who are interested in robots but do not have enough knowledge) to author (create and edit) various motions for the humanoid robot. The new method can derive in real time various motions from the robot postures that users create by turning a dial based on the learning and inference procedures of HTM. Simulation results for authoring a humanoid robot with 26 degrees of freedom (DOF) is given to show the effectiveness of the proposed method.

A State of the Art of Power Transmission Line Maintenance Robots

This paper dealt with a state-of-the-art maintenance robots for power transmission lines (PTL). The paper summarized the PTL maintenance robots that have been developed by eight major research institutions. The main features of the robot navigation were derived and classified through analysis of researches conducted by those institutions. The derived six main features were analyzed quantitatively and qualitatively through readiness level assessment. For this, system readiness level (SRL) was used to evaluate the readiness level of overall system. The SRL was obtained using technology readiness level (TRL) and integration readiness level (IRL). In addition, the readiness level for each feature was reviewed through statistical analysis using box and whisker plot. In conclusion, we confirmed the status and direction of research and development on PTL maintenance robots by analyzing conventional robots with respect to six main features.

Distributed Sensor Network-Based Virtual FDR System

We propose an application for power facility monitoring using distributed sensor networks. The proposed application is a virtual frequency disturbance recorder (FDR) system that estimates the frequency of a power grid using the data from a modeled grid and sends the information to the server for monitoring via wireless communication. To verify the validity of the proposed application, we conducted experiments using actual system data. The functionality of the proposed system was also verified by establishing a ZigBee-based wireless network for power monitoring and diagnosis of disturbances.

A Study on Providing Prompt Assembly Information Using AR Manual

This paper proposes a method for providing prompt assembly information using an augmented reality (AR) manual. Experiments were conducted on widely available personal computer hardware to examine the effectiveness and propriety of the proposed AR manual compared with conventional approaches of using Web and book manuals. The results of the experiments indicate that the performance of the AR manual is much more efficient than that of conventional Web and book manuals.

Automatic motion creating from the posture template for the robot physical training system

This paper presents the collaborated working model to provide service using robot. Besides, we present the physical training system for prototype of the working model. At first in this system, to compare the posture, we define a human-like generic template with two angles, then, adjust the template to fit on each of the person to improve the recognition rate of system. Moreover, we create robot motion that has used angle data of the created template. The angle data is transformed automatically to the most similar motion. This is achieved by the particular expressions that are up to robot specification. The expressions intend to differences of angle measurement on X, Y, Z planes and DOF constraint. We also show the posture recognition and motion transformation are working well.

A Real-Time Video Abstraction Technique for AR Content Synthesis on Android Platform

In this paper, we offer an efficient real-time video abstraction technique that enables interaction with the real-world environment using a smart device for combining augmented reality (AR) and virtual content. The conventional video abstraction technique is inappropriate for the real-time processing of AR content. Therefore, real-time video abstraction through the GPU approach is necessary to minimize the CPU processing of AR content. The methods other than the abstraction technique using the Open GL-ES shader have constraints regarding hardware and the dependency of AR content data. Consequently, this paper proposes a real-time video abstraction technique using the Open GL-ES shader. The proposed real-time video abstraction technique is a method for minimizing the disharmony between virtual and real-world content occurring on AR performing in the limited computing resources of a smart device. To do this, we performed research on video abstraction and applied three filtering algorithms (Sobel operator, Striped pattern-based toon, Non-boundary-based color quantization). In addition, we considered the applicability of AR content through real-time video abstraction experiments using adjusted and monitoring parameters.

An In-Vehicle Application Providing System Based on Driver’s Biodata

The recent technologies for vehicular networks including wireless communication have allowed vehicles to provide biosensor based various applications to a driver. This paper proposes a system to provide a driver with the application which the driver wants in a vehicle using driver’s biodata. The proposed system is composed of four components including a sensing unit, an inference unit, an application providing unit, and vehicular network unit. In this paper, the experiments were performed for correlation investigation between the number of biodata and the reliability of driver’s characteristics extraction. We found that the number of the data used had a significant influence on improving the driver’s satisfaction level, playing an important role in reflecting driver’s status. Experiments using actual vehicles are conducted to examine the effectiveness of the proposed system, and the results are analyzed and discussed. The results on the investigation of detection of biodata in gasoline and diesel vehicles by the experiments are also given. We confirmed that the proposed sensing method was capable of accurate detection irrespective of vehicle mode and vehicle type.

Virtual FDR Based Frequency Monitoring System for Wide-Area Power Protection

There have been recent research activities on GPS-based FNET to prevent wide-area blackouts by monitoring frequency deviation. This paper presented a virtual FDR based monitoring system for monitoring regional frequencies in power grid modeling as an advanced research project for implementing intelligent wide-area protective relaying of South Korea. The system was implemented by modeling an actual 345 kV transmission system using EMTP-RV and by measuring voltages and currents at 5 regions. The frequencies were estimated with a frequency estimation algorithm using gain compensation. The virtual FDR based monitoring system was implemented and simulated in various failure conditions.

A Study on Robot Motion Authoring System using Finger-Robot Interaction

Abstract – This paper proposes a robot motion authoring system using finger-robot interaction. The proposed robot motion authoring tool is a user-friendly method that easily authors creates and controls robot motion according to the number of fingers. Furthermore, the system can be used to simulate user-created robot contents in the 3D virtual environment. This allows the user to not only view the authoring process in real time but also transmit the final authored contents. The effectiveness of the proposed motion authoring system was verified based on motion authoring simulation of an industrial robot. Keywords : Finger-robot interaction, Industrial robot, Robot motion authoring system 1. Introduction With rapid developments in todays robot technology, robots have been used in a wide range of fields. As the increasing number of applications adopts intelligence, and thus robot requires a new type of robot software environment that allows user to author various commands for robots. However, robots must become easier to use for general users in order to expand the scope of applications of intelligent robots in our everyday life. Based on this request, a number of studies are being conducted regarding new types of user-friendly and intuitive robot motion authoring that can render various motions [1]-[4]. Also, there are research projects actively taking place in human-robot interaction, thanks to the progress that has been made in vision technology. A study is being conducted to develop a finger pad that can replace the computer mouse [5], and there are other studies proposing algorithms to improve hand gesture recognition [6]-[8]. However, user-intuitive or user-friendly robot command authoring tools that focus on facilitating general users are still rare. The lack of user-intuitive or user-friendly tools is likely to create a barrier for providing robot services (commands/motions) that correspond with the preferences and demands of general users including children, the elderly and housewives, who are expected to be the major clients in the service robot industry. This paper proposes a robot motion authoring system using finger-robot interaction. The proposed system is capable of authoring (creating and controlling) robot motion according to the number of fingers. It is an intuitive robot motion authoring method that allows the user to easily author robot motions. The effectiveness of the proposed method is verified based on motion authoring simulation of an actual industrial robot.

A Robot Motion Authoring Using Finger-Robot Interaction

This paper proposes a robot motion authoring using finger-robot interaction. The proposed method is a user-friendly method that easily authors (creates and controls) robot motion according to the number of fingers. The effectiveness of the proposed motion authoring method was verified based on motion authoring simulation of an industrial robot.