International Journal of Humanoid Robotics

An adaptive perception system enables humanoid robots to interact with humans and their surroundings in a meaningful context-dependent manner. An important foundation for visual perception is the selectivity of early vision processes that enables the system to filter out low-level unimportant information while attending to features indicated as important by higher-level processes by way of top-down modulation. We present a novel way to integrate top-down and bottom-up processing for achieving such attention-based filtering. We specifically consider the case where the top-down target is not the most salient in any of the used submodalities.

To broadly explore the rationale behind more socially acceptable robot design and to investigate the psychological aspects of social acceptance of robotics, a cross-cultural research instrument, the Robot Assumptions Questionnaire (RAQ) was administered to the university students in Japan, Korea, and the United States, focusing on five factors relating to humanoid and animal-type robots: relative autonomy, social relationship with humans, emotional aspects, roles assumed, and images held. As a result, it was found that (1) Students in Japan, Korea, and the United States tend to assume that humanoid robots perform concrete tasks in society, and that animal-type robots play a pet- or toy-like role; (2) Japanese students tend to more strongly assume that humanoid robots have somewhat human characteristics and that their roles are related to social activities including communication, than do the Korean and the US students; (3) Korean students tend to have more negative attitudes toward the social influences of robots, in particular, humanoid robots, than do the Japanese students, while more strongly assuming that robots' roles are related to medical fields than do the Japanese students, and (4) Students in the USA tend to have both more positive and more negative images of robots than do Japanese students, while more weakly assuming robots as blasphemous of nature than do Japanese and Korean students. In addition, the paper discusses some engineering implications of these research results.

As robots are gradually leaving highly structured factory environments and moving into human populated environments, they need to possess more complex cognitive abilities. They do not only have to operate efficiently and safely in natural, populated environments, but also be able to achieve higher levels of cooperation and communication with humans. Human–robot collaboration (HRC) is a research field with a wide range of applications, future scenarios, and potentially a high economic impact. HRC is an interdisciplinary research area comprising classical robotics, cognitive sciences, and psychology. This paper gives a survey of the state of the art of HRC. Established methods for intention estimation, action planning, joint action, and machine learning are presented together with existing guidelines to hardware design. This paper is meant to provide the reader with a good overview of technologies and methods for HRC.

This paper reports on the development of a multirobot cooperation system for human–robot communication in which robots behave as if communicating by speech, while the system exchanges information through a network that maintains natural communication timing. The information exchanged through the network is based on the analysis of interhuman conversation. This system is based on a scripting language for coordinating multirobot communication, which has a merit for developers: easy development. The developed system is used for Manzai, which is a kind of a Japanese stand-up comedy mainly performed by two persons. While tempo and timing are particularly important in Manzai, the developed "robot Manzai" system received higher evaluations than the Manzai shown in a video performed by humans. We believe that this system demonstrates the potential of robots as a passive–social medium, including television and computers.

Starting from a biomechanical study of the shoulder complex, the relevance of a serial nine d.o.f. kinematic model of the human arm, including a clavicle-like link, was analyzed. It is shown that this partial biomimetic joint model of the upper limb is able to mimic the ability of the natural arm to practically eliminate internal and bound singularities over a large frontal zone, so as to maintain its elbow laterally to the body. In this sense, it appears to be an advanced solution for increasing the dexterity of humanoid robot upper limbs, thus replacing classical seven d.o.f. anthropomorphic arms where a device mimicking the shoulder girdle mechanism is absent.

This work considers some aspects of the problem of generating and preserving of two-legged gait bearing in mind the requirements for a higher degree of similarity with human gait (anthropomorphism) and robustness to the constantly present small disturbances during the walk. Reference motion was synthesized by semi-inverse method, varying the mode of ZMP traveling along a path selected in advance on the foot–ground surface. It was found that different ways of ZMP motion along the path has a decisive role in the trunk deflection in the sagittal plane. Also, the problem of classification of disturbances and their compensation during the gait is discussed in detail. The significance of the multi-link trunk in the gait synthesis and its role in the compensation of disturbances is also considered.