A Review on Robot Motion Legibility for Human-Robot Interaction

Sukyeong Kim,Woojin Park 대한인간공학회 2017 대한인간공학회 학술대회논문집 Vol.2017 No.11

Objective: The purpose of this study was to critically review and synthesize existing studies on robot motion legibility and suggest future directions for human factors research on human-robot interaction (HRI). Three questions were considered: ‘Why is the robot’s legibility important to humans?’, ‘What robot expressions improve robot’s legibility?’, and ‘How does the legibility of robots affect human performance and well-being?’. Background: As robots are quickly becoming an intrinsic part of our daily lives, the demand for technology necessary for the human-robot coexistence beings is increasing. The robot’s legibility is being increasingly recognized as an important design factor for creating human friendly robots. Method: This study critically reviewed and synthesized 30 research articles on the legibility of robot motions. Major findings and limitations of past research studies were examined and suggestions on future human factors research directions were provided. Results: From the psychology point of view, the robot’s legibility has fundamental significance concerning HRI as it is related to the basic human tendency of predicting the goal of a system from its observed behaviors. Also, designing legible robot motions can be understood as creating robot behaviors compatible with human mental models. Many studies have attempted to create highly legible behaviors through approaches, such as imitating human behaviors, considering human abilities, meeting human preferences, and using additional motion cues. Design factors such as actor characteristics, motion trajectory and velocity, were found to affect the legibility of robot. Legible robot motions were found to not only improve the task performance of the whole human-robot collaboration system but also enhance the perceived safety, comfort, trust and positive affect of robots. Conclusion: Although the robot motion legibility is an important human factors research topic concerning HRI, there are currently many research voids that require extensive future research. Application: The current literature review provides an understanding of the recent research trends on the legibility of robot motions and would help human factors researchers define future HRI research problems.

소셜 로봇의 인간-로봇 인터랙션 사례 연구 -소셜 로봇의 디자인 요소를 중심으로-

김나현,박기철 (사)한국커뮤니케이션디자인협회 커뮤니케이션디자인학회 2023 커뮤니케이션 디자인학연구 Vol.82 No.-

The robot market is expanding through software technologies such as AI and machine learning and hardware technologies such as sensors and motors. Among them, home social robots require various social and emotional exchanges with users such as entertainment and companion beyond the function of life assistance. Since this exchange with users is greatly related to the development of human-robot interaction, this case study was conducted to study the method of social interaction. A total of 25 cases of home social robots released or disclosed were analyzed, and for the analysis, the design elements of social robots were set up and the robot interaction analysis framework was set accordingly. The methods of recognition and expression of robots are classified into sound, vision, and others, respectively, and in particular, the robot's expression method was divided into human-like and robot-like methods. In this study, the cases of human-robot interaction were analyzed and will provide as basic research data for a smoother and more efficient way of interaction design in the future. AI, 머신러닝 등의 소프트웨어 기술과 센서, 모터 등 하드웨어의 기술을 통해 로봇 시장은 넓어지고 있다. 그중가정용 소셜 로봇은 생활 보조의 기능을 넘어서 엔터테인먼트, 컴패니언 등 사용자와 다양한 사회적이고 감정적인교류가 요구된다. 사용자와의 이러한 교류는 인간-로봇 인터랙션의 발전과 크게 연관되어 있기에 사회적인 인터랙션의 방식 연구의 기초 연구로 본 사례 연구를 진행하였다. 출시되거나 공개된 총 25종의 가정용 소셜 로봇을 대상으로 분석을 진행하였으며, 분석을 위해 소셜 로봇의 디자인 요소를 설정하고 그에 따른 로봇 인터랙션 분석 프레임워크를 설정하였다. 로봇의 인지와 표현의 방식은 각각 청각, 시각, 기타로 분류되고 특히 로봇의 표현 방식에있어 인간같은(human-like) 방식과 로봇적(robot-like) 방식으로 구분해 분석하고 그 결과로 향후 연구를 위한 디자인 방향성을 제시하였다. 본 연구는 인간-로봇 인터랙션의 사례 연구를 통해 향후 원활하고 효율적인 방식의 인터랙션 디자인의 발전을 위한 기초 연구 자료로 제공될 것을 기대한다.

Human-Robot Cooperative Control Based on pHRI (Physical Human-Robot Interaction) of Exoskeleton Robot for a Human Upper Extremity

이희돈,한창수,이병규,김완수,길명수,한정수 한국정밀공학회 2012 International Journal of Precision Engineering and Vol.13 No.6

This paper proposes a human-robot cooperative control of exoskeleton robot assisting muscle strength of a human upper extremity when lifting or transporting heavy objects. When a human wears a robot, the motions of the human and robot generate interaction, which is called HRI (Human-Robot Interaction). To generate reference motion from the interaction force, a pHRI model was developed using virtual mechanical impedance, and an experimental method to determine the impedance parameters of the pHRI model was proposed. The controller was developed in such a way that the desired motion will be controlled using dynamic model-based compensation. To verify the proposed control method, it was applied to an exoskeleton robot with 6-DOF for both arms. Motion-following-performance experiment and muscle-strengthassisting-effect experiment were conducted using this robot. Experimental results, the wearer of the exoskeleton robot can handle a small force was the heavy object.

Interactive Human Intention Reading by Learning Hierarchical Behavior Knowledge Networks for Human-Robot Interaction

한지형,최승환,김종환 한국전자통신연구원 2016 ETRI Journal Vol.38 No.6

For efficient interaction between humans and robots, robots should be able to understand the meaning and intention of human behaviors as well as recognize them. This paper proposes an interactive human intention reading method in which a robot develops its own knowledge about the human intention for an object. A robot needs to understand different human behavior structures for different objects. To this end, this paper proposes a hierarchical behavior knowledge network that consists of behavior nodes and directional edges between them. In addition, a human intention reading algorithm that incorporates reinforcement learning is proposed to interactively learn the hierarchical behavior knowledge networks based on context information and human feedback through human behaviors. The effectiveness of the proposed method is demonstrated through play-based experiments between a human and a virtual teddy bear robot with two virtual objects. Experiments with multiple participants are also conducted.

지능정보사회에서 지능로봇의 윤리화 과제와 전망 - 근대적 윤리담론에 대한 대안적 접근을 중심으로 -

정채연 전북대학교 동북아법연구소 2018 동북아법연구 Vol.12 No.1

지능정보사회에서 지능로봇의 도입 및 수용에 따라 새로이 제기되는 가치판단의 문제들은 로봇윤리담론의 성장을 이끌고 있다. 로봇윤리는 인공지능 및 지능로봇이 내리는 의사결정이 인간사회에 영향을 미치는 상황을 기본적인 전제로 삼으며, 특히 인간의 생활세계에서 함께 공 존하는 사회적 로봇의 맥락에서 보다 특별한 윤리적 고려가 요청된다고 할 수 있다. 본 논문에서는 로봇윤리담론의 연구성과들을 거시적인 관점에서 조망하면서 논의의 전반적 인 추이 및 동향, 그리고 유의미한 성찰점들을 제시해보고자 한다. 이때 특히 기존의 윤리담론이 상정하고 있는 논의의 전제들을 넘어서거나 적어도 우회하는 대안적 접근에 주목하고자 한다. 이러한 접근은 근대윤리담론에 대한 대안적 성격을 갖는다는 점에서 탈근대적인 방향성을 내포 하고 있다는 평가가 가능할 수도 있겠다. 이를 위해 먼저 로봇윤리의 범주를 세 가지로 구분하고, 로봇윤리에서 주로 다루어지는 지능 로봇의 유형 스펙트럼을 개략적으로 제시한 후, 사회적 로봇이 갖는 독자적 속성 및 특수한 쟁점 을 언급하도록 한다(Ⅱ). 또한 사회적 로봇을 주된 상황적 전제로 하여, 기존의 근대적 윤리담론에 대해 대안적 성격을 갖고 있다고 할 수 있는 로봇윤리의 담론 추이를 다음과 같이 크게 네 가지로 쟁점화한다: i) 자율적인 행위자로서 지능로봇의 윤리적 책무성; ii) 인간과 로봇의 상호작용과 동반자 관계; iii) 인간과 로봇의 정서적 소통과 공감; iv) 로봇을 통한 인간의 역량 강화와 상호진화의 가능성. 이렇듯 로봇윤리담론의 추이에 대한 분석을 통해, 먼저 i) 윤리적 행위자로서 지능로봇을 승인하 기 위하여 기존의 법⋅윤리 개념을 확장적으로 재구성⋅재해석하거나, 인공지능 및 지능로봇에 적합한 새로운 개념들을 창설하려는 시도를 발견할 수 있다. 또한 ii) 인간-로봇 상호작용의 연구 성과에서 잘 나타나듯이 의인화 등 로봇과 관계 맺는 인간의 반응에 주목하면서, ‘정신’에서 ‘신 체’로, 그리고 ‘실재’에서 ‘현상’으로의 관심 전환을 이끌었으며, 이와 더불어 인간-로봇의 관계에 대한 인식이 주종관계에서 동반자관계로 나아가고 있다고 평가할 수 있겠다. 나아가 iii) 인간과 로봇 사이의 상호작용이 갖는 호혜성에 대한 강조는 윤리적 가치판단에 있어서 ‘이성’보다 ‘감정’ 및 ‘공감’의 정서를 중요시하는 논의로 연결된다. 마지막으로 iv) 이러한 소통적 관계는 새로운 기계 종인 로봇과의 상호작용으로 인해 인간 스스로의 자기이해가 변화하고 인간의 전반적인 역량이 강화될 수 있다는 공진화의 가능성에 대한 논의에서 극대화된다(Ⅲ). 인간과 로봇의 상호작용이 인간 자신의 자기이해와 인간적 좋음 및 행복에 기여한다는 윤리 적 논의의 지평은 로봇윤리를 인간윤리의 연장선상에서 이해하도록 하며, 이때 로봇윤리와 휴머 니티와의 양립가능성을 바라볼 수 있게 된다(Ⅳ). In the intelligent information society, several issues raised by artificial intelligence and intelligent robots have led the growth of robot ethics discourse. Ethical issues in robotics are accompanied when a series of decisions made by intelligent robots significantly affect human society. This is especially apparent in the context of social robots interacting and coexisting with human in the real life-world. In this paper, the author takes a view of overall trends of robot ethics discourse and suggests several significant implications for existing legal and ethical theory. Here, the paper focuses on alternative approaches that go beyond or at least bypass the major premises of existing ethical discourse. These approaches could be interpreted as the alternatives to modern ethical theories and therefore inherently having a postmodern directivity. First, the author provides three categories of robot ethics, outlines the types of intelligent robots, and then describes the unique attributes and issues of social robots(II). Furthermore, the paper suggests major alternative approaches to robot ethics discourse as follows: i) ethical accountability of intelligent robots as autonomous agents; ii) human-robot interaction and partnership; iii) emotional communication and empathy between human and robot; and iv) human capabilities and the possibility of co-evolution(III). Finally, as a concluding remark, the author provides future prospects of robot ethics compatible with humanity(IV).

과학관에서의 대중 시연을 통한 인간크기 이족보행 휴머노이드 로봇의 평가 연구

안태범(Tae-Beom Ahn),강이석(E-Sok Kang) 제어로봇시스템학회 2015 제어·로봇·시스템학회 논문지 Vol.21 No.4

Although human-sized bipedal humanoid robots have been developed as the ideal form of human-friendly robots, studies of humanoid robots from the user perspective and of actual interaction between humanoid robots and the public in daily environments are few. This paper presents a long-term public demonstration that encouraged interaction between a humanoid robot and unspecified individuals. We have collected a significant amount of subjective evaluation data from the public by performing a storytelling demonstration that enhanced people’s empathy towards the robot. The evaluation model consists of the robot"s human friendliness, which involves its impression on humans, interaction with humans, and imitation of human motions and the robot"s human appearance which involves gender, age, height, and body type. This study shows that there is no significant difference in human-friendliness between gender groups (male and female), while there is a significant difference between age groups (children and adults). In human appearance, it appears that there is no significant difference between either gender groups or age groups, except for the case of the robot"s height.

휴머노이드 로봇의 얼굴 디자인에 대한 인간의 태도에 관한 연구

정성원(Jeong Seong-Won),이건표(Lee Kun-Pyo) 한국디자인학회 2008 디자인학연구 Vol.21 No.1

For robots to be more useful and supportive to humans, they must be able to respond and communicate accurately to our needs and demands. It is more important that people should understand what robots are saying as much as robot has to know the meaning of human's action; this is the most important thing that many robot designers are considering. If so, how can human beings perceive what robots are saying? In face-to-face interaction, humans communicate relying on mostly their facial expressions, speech and gestures. Among them facial expressions are known to be the most powerful means in conveying one's attitude to others. The present paper is focused on the design of a robot's face that maps out human preferences for the robot's face appearance. The robot’s faces are classified into four groups according to their level of abstractions of appearance. This study conducted a survey with 30 Koreans and 63 Filipinos through questionnaire and short interviews. The questionnaire consists of a series of images that were classified according to the attributes of robot's appearance. To check on the significant difference of human's preferences among groups, one-way ANOVA analysis, Tukey-test and T-test were conducted. The result of the analysis shows significant difference of human's preference for the design of robot faces. For the most part, participants least liked the robots from human-like looking. This study draws the significant disparity between what human likes and familarity to human-likeness. Therefore, the result provides a useful ground to design the robot faces so that robot developers could not be caught in their own snare of human-likeness appearance of robots

소셜로봇 개인화 서비스를 위한 성격 모델링과 표현에 대한 체계적 문헌 검토

김시인(Shee-Ihn Kim),정진우(Jin-Woo Jeong) 한국산학기술학회 2023 한국산학기술학회논문지 Vol.24 No.7

인공지능과 산업용 로봇의 활용으로 산업생산성은 극적으로 향상되어 왔는데, 산업용 로봇과는 달리 소셜로봇은 기능적으로, 사회적으로 용인된 방식으로 행동해야 한다. 그래서 소셜로봇 연구는 언어 표현뿐만 아니라 얼굴표정, 제스처, 행동과 같은 비언어로서 인간과 상호작용을 하는 것으로 확장되어 가고 있다. 소셜로봇의 개인화는 IFTTT(IF This Then That) 방식보다는 로봇에 직접 성격을 부여하는 방향으로 전개되어 가고 있다. 따라서 본 연구에서는 성격검사를 반영하여, 로봇성격, 인간성격이 어떻게 소셜로봇과 인간-로봇 상호작용(Human-Robot Interaction)에 적용되어서, 개인화 서비스를 효과적으로 구현하는 지를 살펴보고자 로봇의 성격, 이를 사용하는 사용자의 성격, 그리고 성격파악에 사용되는 멀티모달의 사회적 단서(Social cues)를 체계적 문헌연구를 통해서 살펴 보았다. 구글 스칼라에서 2018년 이후로 4,220개의 논문 검색하였고, 최종 19편을 선정하였다. 검토결과로 소셜로봇의 성격 모델링은 5편, 사용자 성격 모델링은 16편, 멀티 모달리티 사용은 3편의 논문에서 나타났다. 연도별에서도 성격 모델링과 멀티 모달리티 사용하는 연구가 증가되었고, 인간-로봇 상호작용에 효과적임을 확인하였다. Industrial productivity has improved dramatically with the use of artificial intelligence and industrial robots, but unlike industrial robots, social robots must behave in a functional and socially acceptable way. Therefore, research on social robots is expanding to interact with humans through both verbal expressions and non-verbal expressions, such as facial expressions, gestures, and actions. The personalization of social robots is developing in the direction of giving personalities to robots rather than using if this then that (IFTTT) methods. Therefore, in this study, considering personality tests, we performed a systematic literature to examine the personality of robots, the personality of users, and multimodal social cues used to identify personality. The goal was to examine how robot personality and human personality can be applied to social robots and human-robot interaction to effectively implement personalized services. Google Scholar was searched for relevant papers published since 2018, which yielded 4,220 papers, of which 19 were finally selected. Personality modeling of social robots appeared in 5 papers, user personality modeling appeared in 16 papers, and multimodality usage appeared in 3 papers. Papers using personality modeling and multimodality have increased over time, and their effectiveness in human-robot interaction has been enhanced.

Human-friendly Man-Machine Interaction in Smart Home

Z. Zenn Bien,Jin-Woo Jung,Kwang-Hyun Park,Farrah Wong 한국과학기술원 인간친화 복지 로봇 시스템 연구센터 2005 International Journal of Assistive Robotics and Me Vol.6 No.3

  Independence of people in need in their living environment becomes an imperative notion for any society in the years to come. As an approach to achieve “independence,” this paper addresses the problem of building an assistive home environment emphasizing a human-friendly man-machine interaction. From the functional point of view, a house is considered as a large-scale main (parent) robot which is capable of interacting with residents to provide comfortable living conditions such as appropriate humidity, temperature, lighting, safety and security. Also, under the management of the parent robot, the home system includes several subsystem (children) robots playing their inherent roles to provide the residents with independence and happiness. The way of interaction between the resident and the systems is thus a significant aspect in the design of a smart home for the elderly and people with disability. As an engineering approach, we propose to design based on the idea that the robotic systems should be as human-friendly as possible, and the acquired information in interaction, which includes physical, emotional, mental status and intention of the user, should be incorporated into the system to enhance human-friendliness. After briefly reviewing various aspects of the notion of the human-robot interaction, this paper introduces our experiences of realizing techniques used for human-friendly interaction.

Study of Visual Assist Effect to Vertical Plane Hand Movement during Human-Human Cooperative Task

Shahriman Bin Abu Bakar,Ryojun Ikeura,Yuichiro Handa,Takemi Yano,Hideki Sawai 제어로봇시스템학회 2008 제어로봇시스템학회 국제학술대회 논문집 Vol.2008 No.10

At present the majority of autonomous robots are mostly used in factories where speed and accuracy are given highest priority. In our research, we are focusing in the area where the robot that cooperate with human to lift or carry a human subject. In this area the robots are required to interact with human and move in such a fashion where it will move with human-like motion so that the human subject that is being move will not feel intimidated. In order to design robot that have smooth human like motion capability during human robot interaction in cooperative task, we need to understand how human-human understand each other ,how and what kind of information are exchange between them that enable human-human to be able to accomplish to move object with smooth qualities. Based on this, we need to design a system that is available to be used not only by robotic experts but by general population so that anybody can use this system for their care giving purpose. In this paper we conduct a study of how human-human utilize their sense in moving and stopping an object and we analyzed the smoothness of the motion by analyzing the hand jerk characteristic during the said task.