Published papers

Telepresence robots using traditional video chats do not yet fully represent the presence of a remote user. To enhance that presence, we propose a novel approach that replaces the background of the remote user with a real-time captured image of the area behind the robot. We assumed that the background of the remote user, which was visible on the telepresence robot’s display, was one reason for the reduction in presence because it might remind local viewers that the remote user is in another place. We found that our approach effectively enhanced the presence of remote user and reduced the negative impression of a delay. In this paper, we describe the new approach, its implementation, and an evaluation of the system. We evaluated the subjective impressions of communication through the proposed system. This approach also protects the privacy of the remote user and makes it more acceptable to people using telepresence robots from private places, e.g., a sickroom in a hospital.

In today's era, the security and authenticity of information have always received extensive attention, especially when people have doubts about the authenticity of data. Hence, a traceable data storage platform with the feature that theoretically cannot be tampered with becomes extremely important. Under such demands, the deployment of smart contracts can effectively trace the source and discover problems. This article introduces an environmental data detection platform based on blockchain technology and smart contracts. It aims to record the data continuously and verify the authority of each transaction.

© 2018 IEEE. Inadequate variety of personalities for communication robots may cause unnatural interaction with them and reduction in attachment. This paper proposes C2 AT2 HUB, where communication robots are characterized by changing tendency of affect transition based on long-term interaction with users. In C2 AT2 HUB, robots' affect is defined as two types; 'interpersonal affect' and 'emotions', and transition of each type of affect is adjusted by history of users' actions to robots in order to characterize robots gradually. Evaluation experiment revealed our approach characterizes robots naturally and improves impression of robots.

© 2018 IEEE. In the future, communication robots are expected to join many-to-many human-robot interactions. Thus, robots must handle interruptions requesting a new task outside of the current conversation. In this paper, we propose a novel scheduler which decides switch timing of conversational tasks when a robot is interrupted. The model grasps the structure of the conversation and finds its breakpoints based on adjacency pairs. In order to decide whether to switch conversational tasks on each breakpoint or not, the model prioritizes conversational tasks considering an importance of its topic and a length as contexts of each conversational task. The model also uses human relationships and emotional behaviors to decide priority of conversational tasks. The result of an evaluation experiment shows that our proposed scheduler could impress subjects more favorably than that which always prioritizes an interrupter.

多人数参加型のテレコミュニケーションでは，Fact-to-Faceでの場合と比較して遠隔者の非言語的な振る舞いが伝 わりづらく，会話における発話権の取得が困難，すなわち会話に参加しにくいという問題を抱えていた．この問 題に対し，テレプレゼンスロボットの身体性を強化しアバターロボット化することで，遠隔者の非言語的な振る 舞いを直接的に他の参与者に伝え，発話権の取得を支援する手法が議論されている．しかし，これらの手法では 実際に話者交代が起こるかは受け手の判断に依るため，不確実性が高い．本稿では，仲介者としてより積極的に 遠隔者の発話権の取得を支援するエージェントソフトウェアを搭載したテレプレゼンスロボットの実現を目指し， その第一段階として，遠隔者の会話参与状況と被注視状況の関係性について分析する．

テレプレゼンスロボットは遠隔者の身体的・社会的存在感を再現し，円滑な遠隔コミュニケーションを支援するロボットである.しかし，既存のビデオチャットを用いたロボットでは，完全に遠隔者の存在感が再現されているとはいえない.一因として，画面上における背景の存在がある.背景によって，遠隔者が別の場所に居ることを 明確に意識させてしまうからだ.そこで我々は，ビデオチャットにおける遠隔者映像の背景部分のみをテレプレゼンスロボットが置かれている場の背景映像にリアルタイムに合成することにより，遠隔者の身体的存在感を向上させた ロボットの実装を行い，その評価を行った.

近年3DCGキャラクタが普及し，対話型エージェントのアバタとして画面内エージェントに使用されつつある，しかし画面内エージェントではモナリザ効果というユーザに対する注視が曖昧になる問題が存在する.本研究では会話対象を瞳に映す手法を提案し話者受話者傍観者の分類を行う.またその時の受話者と傍観者に対して，画面内エージェントが適切な対象に会話していたかを実験，検証した.

日常生活において，会話の最中に割込み者が現れて全くコンテキストの異なる会話への参与を求めてくることがしばしば起こる．この割込み場面において，会話参加者らは会話順についての合意形成を行わなければならないが，会話順合意形成タスクは短時間で公正さを追求することが求められるという特徴を持つ．我々は，割込みが発生した際にどのように会話順についての合意形成が行われるかについて分析を目的とした人–ロボット間の会話コーパスの構築を目指している．本論文ではコーパスの設計と分析の指針について議論する．

© 2019 IEEE. The miniaturization and decreasing cost of networking computers, along with the advancement of cloud infrastructure has eased the implementation of IoT products for manufacturers. The increase in such products has led to the rise in IoT devices with information provision function. However, it is difficult to make products display information in consideration of the design on the surface, like daily commodities, since it requires the device to have a wealth of computational resource. In this paper, we propose Virtual IoT System, the approach which adds information-giving functions to the non-computing objects using Mixed Reality technology. This system enables appropriate things to give adequate information at the appropriate timing. In this paper, we explain the detail and design of Virtual IoT System. In addition, we conduct an experiment to evaluate the usability and expressive power of Virtual IoT system.