International Journal of Humanoid Robotics

Characters that cross dimensions have elicited an avid interest in literature and cinema. In analogy to these characters, we explore the concept of migration: Process by which an agent moves between embodiments, being active in only one at a time. We developed an autonomous artificial pet with two embodiments: A virtual within a smartphone and a physical robotic embodiment. Considering that owners' interactions with real pets lead to emotional attachment and potentially related health benefits, we conducted a user study with elementary school students to assess their attachment to the prototype and how natural they felt the interaction was. By the end of the experiment children felt closer to the artificial pet and 43.3% considered the two embodiments to correspond to the same entity, although migration was never explained to them. As a result, this paper presents a novel generic methodology that allows the evaluation of other implemented prototypes that support migration. Furthermore, we created a set of design guidelines for migrating agents.

It is challenging to recognize human hand grasp movements through electromyogram (EMG) signals. In this paper, EMG–EMG coherence measure is used to describe correlations of EMG recordings during different hand grasp movements. Meanwhile, linear discriminant analysis (LDA) is utilized to evaluate the performance of the EMG–EMG coherence measure for identifying the grasp movements. Experimental results show that the EMG–EMG coherence measure is effective to extract correlations among EMG recordings, with which different types of hand grasp movements have been successfully distinguished. It is shown that the EMG–EMG coherence measure might be a potential tool to reveal the EMG–EMG correlation between the intermuscular interactions during hand grasp movements.

This paper investigates the role of interaction and communication kinesics in human–robot interaction. This study is part of a novel research project on sign language (SL) tutoring through interaction games with humanoid robots. The main goal is to motivate the children with communication problems to understand and imitate the signs implemented by the robot using basic upper torso gestures and sound. We present an empirical and exploratory study investigating the effect of basic nonverbal gestures consisting of hand movements, body and face gestures expressed by a humanoid robot, and having comprehended the word, the participants will give relevant feedback in SL. This way the participant is both a passive observer and an active imitator throughout the learning process in different phases of the game. A five-fingered R3 robot platform and a three-fingered Nao H-25 robot are employed within the games. Vision-, sound-, touch- and motion-based cues are used for multimodal communication between the robot, child and therapist/parent within the study. This paper presents the preliminary results of the proposed game tested with adult participants. The aim is to evaluate the SL learning ability of participants from a robot, and compare different robot platforms within this setup.

Multi-degrees of freedom (DOF) parallel robot, due to its compact structure and high operation accuracy, is a promising candidate for medical rehabilitation devices. However, its controllability relating to the nonlinear characteristics challenges its interaction with human subjects during the rehabilitation process. In this paper, we investigated the control of a parallel robot system using fuzzy sliding mode control (FSMC) for constructing a simple controller in practical rehabilitation, where a fuzzy logic system was used as the additional compensator to the sliding mode controller (SMC) for performance enhancement and chattering elimination. The system stability is guaranteed by the Lyapunov stability theorem. Experiments were conducted on a lower limb rehabilitation robot, which was built based on kinematics and dynamics analysis of the 6-DOF Stewart platform. The experimental results showed that the position tracking precision of the proposed FSMC is sufficient in practical applications, while the velocity chattering had been effectively reduced in comparison with the conventional FSMC with parameters tuned by fuzzy systems.

Visual-based human action recognition is currently one of the most active research topics in computer vision. The feature representation directly has a crucial impact on the performance of the recognition. Feature representation based on bag-of-words is popular in current research, but the spatial and temporal relationship among these features is usually discarded. In order to solve this issue, a novel feature representation based on normalized interest points is proposed and utilized to recognize the human actions. The novel representation is called super-interest point. The novelty of the proposed feature is that the spatial-temporal correlation between the interest points and human body can be directly added to the representation without considering scale and location variance of the points by introducing normalized points clustering. The novelty concerns three tasks. First, to solve the diversity of human location and scale, interest points are normalized based on the normalization of the human region. Second, to obtain the spatial-temporal correlation among the interest points, the normalized points with similar spatial and temporal distance are constructed to a super-interest point by using three-dimensional clustering algorithm. Finally, by describing the appearance characteristic of the super-interest points and location relationship among the super-interest points, a new feature representation is gained. The proposed representation formation sets up the relationship among local features and human figure. Experiments on Weizmann, KTH, and UCF sports dataset demonstrate that the proposed feature is effective for human action recognition.

Implementation of robotic writing ability is recognized as a difficult task, which involves complicated image processing and robotic control algorithms. This paper introduces a novel approach to robotic writing by using human–robot interactions. The method applies a motion sensing input device to capture a human demonstrator's arm trajectories, uses a gesture determination algorithm to extract a Chinese character's strokes from these trajectories, and employs noise filtering and curve fitting methods to optimize the strokes. The approach displays real-time captured trajectories to the human demonstrator; therefore, the human demonstrator is able to adjust his/her gesture to achieve a better character writing effect. Then, our robot writes the human-gestured character by using the robotic arm's joint values. The inverse kinematics algorithm generates the joint values from the stroke trajectories. Experimental analysis shows that the proposed approach can allow a human to naturally and conveniently control the robot in order to write many Chinese characters. Additionally, this approach allows the robot to achieve a satisfactory writing quality for characters with a simple structure, with the potential to write more complex characters.

Robot skills are motion or grasping primitives from which a complicated robot task consists. Skills can be directly learned and recognized by a technique named programming-by-demonstration. A human operator demonstrates a set of reference skills where his motions are recorded by a data-capturing system and modeled via fuzzy clustering and a Takagi–Sugeno modeling technique. The skill models use time instants as input and operator actions as outputs. In the recognition phase, the robot identifies the skill shown by the operator in a novel test demonstration. Finally, using the corresponding reference skill model the robot executes the recognized skill. Skill models can be updated online where drastic differences between learned and real world conditions are eliminated using the Broyden update formula. This method was extended for fuzzy models especially for time cluster models.

A fast method based on binocular vision is proposed for mobile robot to detect drivable regions. At first, the image is segmented into regions, and some obstacles are determined by the ground vanishing line. Then, according to the different distribution of feature points extracted from the left regions, we propose two approaches to classify regions: region determination based on feature statistical classification for regions with rich feature points and region determination based on area statistical classification for regions with sparse feature points. Finally, we get the drivable regions by combination of the two approaches. The results of indoor experiments show that the method can perform quickly and robustly.

Visual-based action recognition has already been widely used in human–machine interfaces. However, it is a challenging research to recognize the human actions from different viewpoints. In order to solve this issue, a novel multi-view space hidden Markov models (HMMs) algorithm for view-invariant action recognition is proposed. First, a view-insensitive feature representation by combining the bag-of-words of interest point with the amplitude histogram of optical flow is utilized for describing the human action sequences. The combined features could not only solve the problem that there was no possibility in establishing an association between traditional bag-of-words of interest point method and HMMs, but also greatly reduce the redundancy in the video. Second, the view space is partitioned into multiple sub-view space according to the camera rotation viewpoint. Human action models are trained by HMMs algorithm in each sub-view space. By computing the probabilities of the test sequence (i.e., observation sequence) for the given multi-view space HMMs, the similarity between the sub-view space and the test sequence viewpoint are analyzed during the recognition process. Finally, the action with unknown viewpoint is recognized via the probability weighted combination. The experimental results on multi-view action dataset IXMAS demonstrate that the proposed approach is highly efficient and effective in view-invariant action recognition.