Search

In this paper, we investigate which approach to generate eye behaviors using an android robot makes what impressions on humans and clarify which are the important factors for attractive eye behaviors. Thus, we evaluate the human impression of eye behaviors displayed by an android robot while talking to a human by comparing the motions generated by the two approaches. In the first approach, we develop a method to imitate human eye behavior obtained from eye trackers. In the second approach, we control the eye direction, eye-contact duration, and eyeblinks based on the findings of human eye behavior in psychology and cognitive research. In the experiments, we asked male and female subjects to evaluate their impression by comparing the eye behaviors with an android that controls the eye-contact duration and eyeblinks by editing the imitation parameters or the rule-based behavior. In the experiments, we asked subjects to evaluate their impression of different eye behaviors displayed by an android. The eye behaviors were generated by modifying the imitation parameters or the rule-based behavior, which resulted in different eye-contact duration and eyeblink duration and timing. From the results, we found that (1) the imitation and rule-based behaviors showed no difference in terms of human-likeness, (2) the 3-second eye contact obtained better scores regardless of the imitation or rule-based eye behavior, (3) the subjects might regard the long eyeblinks as voluntary eyeblinks, with the intention to break eye contact, and (4) female subjects preferred short eyeblinks rather than long ones and considered that short eyeblinks might be one of the keys to making eye contact more suitable, in contrast to male subjects who preferred long eyeblinks. [ABSTRACT FROM AUTHOR]

This paper presents a new framework to synthesize humanoid behavior by learning and imitating the behavior of an articulated body using motion capture. The video-based motion capturing method has been developed mainly for analysis of human movement, but is very rarely used to teach or imitate the behavior of an articulated body to a virtual agent in an on-line manner. Using our proposed applications, new behaviors of one agent can be simultaneously analyzed and used to train or imitate another with a novel visual learning methodology. In the on-line learning phase, we propose a new way of synthesizing humanoid behavior based on on-line learning of principal component analysis (PCA) bases of the behavior. Although there are many existing studies which utilize PCA for object/behavior representation, this paper introduces two criteria to determine if the dimension of the subspace is to be expanded or not and applies a Fisher criterion to synthesize new behaviors. The proposed methodology is well-matched to both behavioral training and synthesis, since it is automatically carried out as an on-line long-term learning of humanoid behaviors without the overhead of an expanding learning space. The final outcome of these methodologies is to synthesize multiple humanoid behaviors for the generation of arbitrary behaviors. The experimental results using a humanoid figure and a virtual robot demonstrate the feasibility and merits of this method. [ABSTRACT FROM AUTHOR]

It is a formidable issue how robots can show behaviors to be considered as the corresponding human's ones since the body structure is different between robots and humans. As a simple case for such a correspondence problem, this paper presents a robot that learns to vocalize vowels through the interaction with its caregiver. Inspired by the findings of developmental psychology, we focus on the roles of maternal imitation (i.e., imitation of the robot voices by the caregiver) since it could play a role to instruct the correspondence of the sounds. Furthermore, we suppose that it causes unconscious anchoring in which the imitated voice by the caregiver is performed unconsciously, closely to one of his/her own vowels and hereby helps to make the robot's utterances to be more vowel-like. We propose a method for vowel learning with an imitative caregiver, under the assumptions that the robot knows the desired categories of the caregiver's vowels, and the rough estimate of mapping between the region of sounds that the caregiver can generate and the region that the robot can. Through experiments with a Japanese imitative caregiver, we show that a robot succeeds in acquiring more vowel-like utterances than a robot without such a caregiver, even when it is given different types of mapping functions. [ABSTRACT FROM AUTHOR]

This paper describes a humanoid robot system that can capture and mimic the motion of human body parts in real-time. The underlying vision system is able to automatically detect and track human body parts such as hands and faces in images captured by the robot's eyes. It is based on a probabilistic approach that can detect and track multiple blobs in a 60-Hz stereo image stream on a standard dual processor PC. A random jerk model is employed to approximate the observed human motion and a Kalman filter is used to estimate its parameters (three-dimensional positions, velocities and accelerations). This enables the system to realistically mimic the perceived motion in real-time. [ABSTRACT FROM AUTHOR]

A method to learn and reproduce robot force interactions in a human-robot interaction setting is proposed. The method allows a robotic manipulator to learn to perform tasks that require exerting forces on external objects by interacting with a human operator in an unstructured environment. This is achieved by learning two aspects of a task: positional and force profiles. The positional profile is obtained from task demonstrations via kinesthetic teaching. The force profile is obtained from additional demonstrations via a haptic device. A human teacher uses the haptic device to input the desired forces that the robot should exert on external objects during the task execution. The two profiles are encoded as a mixture of dynamical systems, which is used to reproduce the task satisfying both the positional and force profiles. An active control strategy based on task-space control with variable stiffness is then proposed to reproduce the skill. The method is demonstrated with two experiments in which the robot learns an ironing task and a door-opening task. [ABSTRACT FROM AUTHOR]