Papers - OSAWA Hirotaka

FAN-14-017 Designing Protocol of Werewolf Game : Protocol for Inference and Persuasion

OSAWA Hirotaka, TORIUMI Fujio, KATAGAMI Daisuke, SHINODA Kosuke, INABA Michimasa

FAN Symposium : Intelligent System Symposium-fuzzy, AI, neural network applications technologies (The Japan Society of Mechanical Engineers)  2014 ( 24 ) 78 - 81 2014

Joint Work

FAN-14-016 A proposal of Ai Werewolf Game Challenge for Standard AI Problem

Shinoda Kosuke, Toriumi Fujio, Inaba Masamichi, Osawa Hirotaka, Katagami Daisuke

FAN Symposium : Intelligent System Symposium-fuzzy, AI, neural network applications technologies (The Japan Society of Mechanical Engineers)  2014 ( 24 ) 74 - 77 2014

Joint Work

Emotional cyborg: human extension with agency for emotional labor.

Hirotaka Osawa

ACM/IEEE International Conference on Human-Robot Interaction(HRI) (ACM)     108 - 108 2014

Joint Work, Lead author, Last author, Corresponding author, Accepted

• Access to Document (DOI)

Emotional cyborg: complementing emotional labor with human-agent interaction technology.

Hirotaka Osawa

Proceedings of the second international conference on Human-agent interaction(HAI) (ACM)     51 - 57 2014

Joint Work, Lead author, Last author, Corresponding author, Accepted

• Access to Document (DOI)

Digital play therapy for children with learning disabilities.

Yukako Watanabe, Yoshiko Okada, Hirotaka Osawa, Midori Sugaya

Proceedings of the second international conference on Human-agent interaction(HAI) (ACM)     185 - 188 2014

Joint Work, Accepted

• Access to Document (DOI)

Sustainable relationship with product by implementing intentional interaction.

Hirotaka Osawa

International conference on Autonomous Agents and Multi-Agent Systems(AAMAS) (IFAAMAS/ACM)     1677 - 1678 2014

Joint Work, Lead author, Last author, Corresponding author, Accepted

The Simulation of Bullying and Conforming at a Class Based on Socion Theory

Osumi Toshihiro, Osawa Hirotaka, Imai Michita

IEEJ Transactions on Electronics, Information and Systems (The Institute of Electrical Engineers of Japan)  134 ( 4 ) 560 - 570 2014

Joint Work, Accepted,  ISSN  0385-4221

　View Summary

Bullying is a serious problem, especially at elementary school and junior high school and occurs all over the world with some differences. As a countermeasure against bullying, many interventions and preventive programs were proposed (e.g., counseling, dispatch of advisers, group work, suspension order) and there are many researches about analysis of bullying and intervention and preservation strategies. However, a lot of samples and long times are required for analysis of the strategies in the real world. And we believe simulations of the bullying and the countermeasures are effective as a quick and easy analysis. In this paper, we developed a simulation model of school bullying "within-group" or "within-class" that is based on "Socion theory". The primary feature of "Socion theory" is that each agent has an imagined personal relationship. We implemented a strategy named conforming strategy to simulate the bullying caused by conforming behaviors of students. The student agents who implement the conforming strategy watch the human relationship around them and update their own imagined relationships. Then they conform to other students to get good positions or keep their grounds on the basis of simulation results based on their own imagined relationships. The result of the experiment shows the conforming strategy increases the probability of bullying and the number of victims, and conforming behaviors increase the probability of bullying in the real world. We also show a state that a teacher underestimates or cannot perceive bullying when it occurred, and many teachers underestimate or cannot perceive bullying in the real world. Totally, we show that "Socion theory" has expressiveness to use as a foundation of bullying simulation.

• Access to Document (DOI)
• CiNii Research

Function Learning by Self-explanatory Interface

OSAWA Hirotaka

電子情報通信学会技術研究報告. CNR, クラウドネットワークロボット (The Institute of Electronics, Information and Communication Engineers)  113 ( 84 ) 9 - 12 2013.06

Single Work, Lead author, Last author, Corresponding author,  ISSN  0913-5685

　View Summary

Self-explanatory interface helps self-learning process of learner without third person explainer between learner and learned appliance. The learner need not to conscious about explainer during the learning process. This method sustains self-respect of each learner and continues motivation for learning. In this study, we proposes self-explanatory interface that changes the level of appearance according to user's attribute. We implemented several hardware to control anthropomorphic presence of home-appliance for supporting self-explanatory learning method.

触知覚伝達は人とロボットに一体感をもたらすか?

金井 祐輔, 大澤 博隆, 今井 倫太

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 (東京 : 電子情報通信学会)  113 ( 84 ) 19 - 23 2013.06

ISSN  09135685

Research Trends on Human-Agent Interaction(<Special Issue>Agent Technology)

OSAWA Hirotaka, Hirotaka Osawa

Journal of the Japanese Society for Artificial Intelligence (The Japanese Society for Artificial Intelligence)  28 ( 3 ) 405 - 411 2013.05

Joint Work, Lead author, Last author, Corresponding author,  ISSN  0912-8085

• Access to Document (DOI)
• CiNii Research

Merging Viewpoints of User and Avatar in Telecommunication Using Image and Sound Projector

大澤,博隆, 野田,誠人, 大村,廉, 今井,倫太

情報処理学会論文誌 54 ( 4 ) 1413 - 1421 2013.04

Joint Work, Accepted,  ISSN  1882-7764

　View Summary

本論文では，仮想的な身体を持つアバタを投影する遠隔コミュニケーションシステムPROT AVATARにおけるアバタ操作手法に関する実験をもとに，得られた知見について議論する．PROT AVATARによるコミュニケーションでは，アバタの操作者の映像を遠隔地に投影するため，表情により感情を伝えることができる．さらに，アバタの操作者にとっては明確ではない，アバタの投影に適切な位置をシステムが自動で計算するため，アバタの操作者は投影位置を考えることなく遠隔の環境内を指し示すことができる．しかし，アバタの操作者が採用する視点はアバタの視点とは異なることがあるため，アバタの操作者の発話がアバタとの対話者にとっては自然ではない場合がある．本論文では，アバタの操作手法として，自動操作手法と半自動操作手法の2つの手法を設計・実装し，比較実験を行った．実験の結果，半自動操作手法のほうが自動操作手法よりも，アバタとの対話者にとって自然な発話を引き出すことが示された．また，実験を通して得られた遠隔コミュニケーションシステム関する知見をまとめる．This paper discusses the findings of the viewpoint of an avatar-controlling user on the basis of experimentation with an implemented telecommunication system named PROT AVATAR. Communication using an avatar with facial expressions is useful when a user wants to express emotions. On top of this feature, our system supports avatar movement toward nearest visible location to the target, which is not obvious for the avatar controller. With our system, the avatar controller can easily refer to something remotely. However, sometimes, the words of an avatar controller may not be intuitive for an avatar viewer, because the avatar controller does not necessarily share the viewpoint of the avatar. We designed automatic and semi-automatic methods for controlling the avatar, and we conducted an experiment to compare the two methods. The results showed that semi-automatic control brought more intuitive utterances for an avatar viewer than fully automatic control, and they have design implications for telecommunication systems.

Merging Viewpoints of User and Avatar in Telecommunication Using Image and Sound Projector

石井,健太郎, 谷口,祐司, 大澤,博隆, 中臺,一博, 今井,倫太

Journal of information processing 54 ( 4 ) 1413 - 1421 2013.04

Joint Work, Accepted

全方位台車を用いたロボットアバタシステム

中谷 慶太, 大澤 博隆, 今井 倫太

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 (東京 : 電子情報通信学会)  112 ( 436 ) 27 - 30 2013.02

ISSN  09135685

Collecting Task between Human and Empty-box using Behaviors with Two-axis Actuators

大澤,博隆, 飛田,国星, 桑山,裕基, 今井,倫太, 山田,誠二

The transaction of Human Interface Society (ヒュ-マンインタフェ-ス学会)  15 ( 1 ) 39 - 50 2013.02

Joint Work, Lead author, Corresponding author, Accepted,  ISSN  1344-7262

　View Summary

<p>In this paper, we propose to make intention drawing from system to human and collecting task between human and empty-box just by behaviors with two axis actuators. First, we extracted elements of minimum interaction model between user and empty-box using immersion discovering method based on Wizard of Oz method. Next, we designed three-state model based on findings. Third, we implemented three-state model by motion capture system, and confirms how collecting task was achieved. Last, we evaluated these models are indistinguishable from human manipulation using Turing-test based evaluation. </p>

• Access to Document (DOI)
• CiNii Research

分離した身体を持つロボットによるインタラクション戦略の生成

大澤, 博隆, Wongphati, Mahisorn, 今井, 倫太

Proc. of JSAI (The Japanese Society for Artificial Intelligence)  2013   3J1OS20a2 - 3J1OS20a2 2013

Joint Work, Lead author, Corresponding author

　View Summary

<p>著者らは、人間と異なるロボットの意図伝達戦略を発見する手法として、人間の操作者がロボットに成り代わりながらインタラクション戦略を発見する没入型発見法を提案している。本研究では没入型発見法を使用し、回転台とロボットアームという複数のロボットを操作者が自分の身体として操り、積み木の組み立て方を参加者に対し教える実験を行った。分析結果を通して、参加者がどのように意図を理解したか検討する。</p>

• Access to Document (DOI)

交代取引ゲームにおける他者識別規則の進化

大澤, 博隆, 今井, 倫太

Proc. of JSAI (The Japanese Society for Artificial Intelligence)  2013   4H13 - 4H13 2013

Joint Work, Lead author, Corresponding author

　View Summary

<p>交代取引ゲームは反復囚人のジレンマゲームの一種であり、相互協調よりも協調、裏切りの順番を交代する際に両者の最終利得が高くなる。本研究では交代取引ゲームを人工生命によるシミュレーションと人間集団同士の交代取引ゲームの両者から調べた。結果、交代取引ゲームにおいて初手の交渉結果をシグナルとして固有の信号を発信し、交渉に価する他エージェントを識別する他者信頼能力が発達していくことがわかった。</p>

• Access to Document (DOI)

Blended Realityにより遠隔コミュニケーションを支援するアバターの設計

金井, 祐輔, 大澤, 博隆, 今井, 倫太

Proc. of JSAI (The Japanese Society for Artificial Intelligence)  2013   1M4OS18b4 - 1M4OS18b4 2013

Joint Work

　View Summary

<p>本稿ではblended realityにより遠隔地に居る人とのコミュニケーションを支援するアバターを提案する。Blended realityとは実世界空間と仮想空間を連続的に繋げ、両空間間でのインタラクションを可能にする技術をいう。Blended reality空間でインタラクションすることで実世界空間インタラクションと仮想空間インタラクションの両方の利点を生かしたインタラクションを可能にする。</p>

• Access to Document (DOI)

Pygmy: A ring-shaped robotic device for storytelling

Ogata, M., Sugiura, Y., Osawa, H., Imai, M.

International Journal of Innovative Computing, Information and Control 9 ( 12 )  2013

Joint Work, Accepted

Morphing agency: deconstruction of an agent with transformative agential triggers.

Hirotaka Osawa, Michita Imai

Annual conference on Human Factors in Computing Systems (CHI) (ACM)     2237 - 2246 2013

Joint Work, Lead author, Corresponding author, Accepted

• Access to Document (DOI)

Interaction with an agent in blended reality.

Yusuke Kanai, Hirotaka Osawa, Michita Imai

PROCEEDINGS OF THE 8TH ACM/IEEE INTERNATIONAL CONFERENCE ON HUMAN-ROBOT INTERACTION (HRI 2013) (IEEE)     153 - 154 2013

Joint Work, Accepted,  ISSN  2167-2121

　View Summary

This paper proposes a Blended Reality Agent called "BReA" that can exist in both the real and virtual world and has communicative advantages that is more than simply the sum of those of a robotic agent and an on-screen agent. BReA can seamlessly transfer between the real world and virtual world and users can communicate with BReA without interruption during the transference. We conducted a field test and showed that users can recognize when it points from the virtual world to a real world object after it moves from the real world to virtual world. This result supports a hypothesis that communicating with BReA helps users recognize the link between real world and virtual world objects and virtual world information.