Woong Hee Shon

Development of an android for singing with facial expression

This paper presents the hardware mechanism and software architecture of a singer robot system called EveR-2. EveR-2 is an android robot platform has human-like appearance and shows its emotion with facial expression and gestures. The skin of the head, arms, hands and legs is made of silicon jelly to give human-like texture. EveR-2 has sixty-two degrees of freedom in the head, neck, arms, hands, torso, and legs. It sings a song by reading a music score with lip synchronization. EveR-2 is the first android that made her debut and a music video.

A behavior combination generating method for reflecting emotional probabilities using simulated annealing algorithm

This paper presents a behavior generating method for reflecting emotional probabilities. The proposed method consists of two processes: an emotion-behavior probability generating process and a unit behavior combination generating process. 1) In the emotion-behavior probability generating process, the emotional probabilities of behaviors are determined on the basis of user preferences in terms of the priorities of emotions. 2) In the unit behavior combination generating process, optimal behaviors are found by the simulated annealing algorithm. A final behavior is a set of selected parts of expressions. It is possible to not only reveal an abundance of expressions without one-to-one mapping relations between emotions and behaviors but also apply these expressions in the case of various robots. We have verified the diversity of emotional expression by applying the proposed method to two different robot systems, which are a cyber robot simulator and a real robot system.

Natural changing of emotional expression by considering correlation of behavior history

This paper presents a novel method for natural changing of emotional expression. Although formal behavior generation method using the simulated annealing algorithm selects various behaviors, there are weak correlations among behavior history. To make strong correlations, we restrict the randomness of the simulated annealing algorithm. This method sifts unapt candidate behavior combinations out by considering the previous behavior combinations and makes adequate range of candidate behavior combinations. We have verified the continuity of the proposed method using a cyber robot simulator.

Stackable manipulator for mobile manipulation robot

This paper proposes a new manipulator concept applied to a mobile robot manipulation system for reducing robot size and weight or increasing its work capacities such as a payload, operating radius, and operating speed. In detail, we propose a new robotic manipulator that uses stackable 4-BAR mechanisms for mobile manipulation robot. The proposed mechanism provides a clear advantage in which all the actuators can be separated from the working joints. Thus, the mechanism is able to select the Center of Mass (CoM) and the Zero-Moment Point (ZMP) in arbitrary points without any support from ZMP controller or ZMP compensation method. To confirm efficiency of the new manipulator, this paper addresses a design method using the simplified beam theory, based on the well-known Finite Element Method (FEM) for structural stiffness analysis of linkages. The reason behind this is that the CoM and ZMP are dependent on the weight of the motors and the linkages. Ultimately, we show the efficiency of the proposed stackable manipulator through simulations and experiments.

Design of a personalized r-learning system for children

This paper introduces the r-learning (robot-learning) system that utilizes robotic interactions to personalize instructions for individual children. A child who is not familiar with using an educational tool to study could learn readily from educational contents through various interactions with a robot which is based on the behavior control. The contents which are chosen according to a childs learning data extracted by analysis from human-robot interaction data could lift the educational effectiveness. An r-learning scenario is designed by using Petri net to handle exception occurred in a learning process. The robot contents maker and the Open API for operating a robot are described as management utilities for a personalized r-learning system.

Preliminary thruster control experiments for underwater vehicle positioning

To improve the maneuverability of underwater vehicles, accurate thruster control is required based on a precise thrust model. Especially, in real systems, the ambient flow velocity and angle effects on the thrust force are very important, because vehicles are continuously moving and/or hovering with thruster actions. In this paper, we execute preliminary experiments with the model containing ambient flow effects as CAR (critical advance ratio) and CIA (critical incoming angle). The distinguished characteristics of the above model is the three axial flow states classification. The whole thrust map is divided into three states according to the state of ambient flow velocity, angle, and propeller shaft velocity. The several controllers are tested using the three flow states information. The preliminary experimental results show the best performance can be obtained by the open loop control with accurate model, because the thrust force cannot be measured directly, so the force map from the propeller shaft velocity to thrust force plays important roll in control performance. However, the DOB (disturbance observer) with closed loop controller is realistic solution, because it can compensate various unknown disturbances in real systems