Showing papers in "Journal of robotics and mechatronics in 2008"

Development of a Handshake Robot System Based on a Handshake Approaching Motion Model

Abstract: Handshake refers to embodied interaction using physical contact for closeness. In this paper, we analyze the handshake approaching motion between humans in cases with and without voice greeting and propose a handshake approaching motion model based on the analysis. In this model, a robot can generate handshake approaching motion that is emotionally acceptable to humans by using the second-order lag element and the dead time element from trajectory of human hand. Furthermore, a handshake robot system that adopts this proposed model is developed and motion characteristics of handshake approaching motions with and without voice greeting are analyzed using analysis-by-synthesis technique. The effectiveness of this model and the handshake robot system are demonstrated by sensory evaluation.

Origami Folding by a Robotic Hand

Abstract: For the purpose of realizing the functions of the human hand on a mechanical system, fundamental understanding of the human hand is necessary through intensive observations of manipulation tasks by humans. As an example of skilled manipulation tasks, we target origami folding. We first analyzed the difficulty of origami manipulation and observed how the human folds the origami. We then designed a robotic mechanism that can make an origami work, the Tadpole, in the same way as the human does based on the observation of the origami folding by the human. Although the appearance of the developed robotic hand is dissimilar to that of the human hand, the robot succeeded in performing the origami task, the Tadpole.

Wearable Master-Slave Training Device for Lower Limb Constructed with Pneumatic Rubber Artificial Muscles

Abstract: The purpose of this study is to develop a wearable master-slave lower limb training device for a trainer and a trainee. The developed device is constructed with McKibben type rubber artiacial muscles and an appliance, a torque sensor. In addition, to prevent applying an excessive torque to the trainee, a reaction torque from the trainee can be transmitted to the trainer by a bilateral type master-slave control system. In this paper, the structure of the developed device is discussed, and then the validity of the proposed device is evaluated from the experiments assumed a training.

Learning of Joint Attention from Detecting Causality Based on Transfer Entropy

Abstract: Joint attention, i.e., the behavior of looking at the same object that another person is looking at, plays an important role in human and human-robot communication. Previous synthetic studies focusing on modeling the early developmental process of joint attention have proposed learning methods without explicit instructions for joint attention. In these studies, the causal structure between a perception variable (a caregiver’s face direction or an individual object) and an action variable (gaze shift to a caregiver’s face or to an object location) was given in advance to learn joint attention. However, such a structure is expected to be found by the robot through interaction experiences. In this paper, we investigates how transfer entropy, an information theory measure, is used to quantify the causality inherent in face-to-face interaction. In computer simulations of human-robot interaction, we examine which pair of perceptions and actions is selected as the causal pair and show that the selected pairs can be used for learning a sensorimotor map for joint attention.

Adaptive Gait for Large Rough Terrain of a Leg-wheel Robot (First Report : Gait Strategy)

Abstract: A leg-wheel robot with four mechanically separated legs and two wheels is highly mobile and stable on rough terrain. We discuss the strategy for the robot movement over large rough terrain, classifying topographical features into 13 patterns of combined terrain surface. To traverse all classified terrain, we propose three adaptive gaits: (1) Step-up gait in which frontfoot landing is higher than contact with the wheel ground, and the robot raises itself toward frontfoot landing; (2) Step-down gait in which frontfoot landing is lower than contact with the wheel ground, and the robot lowers itself toward frontfoot landing; and (3) Step-over gait in which frontfoot landing is no higher than contact with the wheel ground, but the robot raises itself as high as possible. 1

Hierarchical Implicit Feedback Structure in Passive Dynamic Walking

Abstract: The purpose of this paper is to analyze the stability of Passive Dynamic Walking (PDW) using a linearized analytical Poincare map. In particular, in this paper, we focus on a bifurcation phenomenon in PDW. Although the bifurcation of the walking period is one of the well-known features of PDW, it have not been studied sufficiently so far. Using techniques similar to our previous research, we derive an analytical Poincare map for 2-period walking and discuss the stability of PDW with this map. In addition, we point out that there is a similar interesting structure in this Poincare map.

Feed-Forward Controller for the Integrated Non-Invasive Ultrasound Diagnosis and Treatment

Abstract: The integrated non-invasive ultrasound diagnosis andtreatment we propose tracks and follows movementin an affected area –kidney stones here– while High-Intensity Focused Ultrasound (HIFU) is irradiatedonto the area. High-speed CCD camera cannot beused in non-invasive diagnosis and treatment becausewe must avoid damaging healthy tissue. Servoing er-ror mainly due to ultrasound imaging and its deadtime become serious problems, unlike when a high-speed camera is used. We propose feed-forward con-trol using semi-regular kidney movement focusing onenhancing servoing performance.Keywords: non-invasive ultrasound diagnosis and treat-ment, motion tracking, feed-forward control, High Inten-sity Focused Ultrasound, medicalsupport system

Constructive Approach to Role-Reversal Imitation Through Unsegmented Interactions

Abstract: This paper presents a novel method of a robot learning through imitation to acquire a user’s key motions automatically. The learning architecture mainly consists of three learning modules: a switching autoregressive model (SARM), a keyword extractor without a dictionary, and a keyword selection filter that references to the tutor’s reactions. Most previous research on imitation learning by autonomous robots targeted motions given to robots, were segmented into meaningful parts by the users or researchers in advance. To imitate certain behavior from continuous human motion, however, robots must find segments to be learned. To achieve this goal, the learning architecture converts a continuous time series into a discrete time series of letters using the SARM, finds meaningful segments using the keyword extractor without a dictionary, and removes less s meaningful segments from keywords using the user’s reactions. In experiments, an operator showed unsegmented motions to a robot, and reacted to the motions the robot had acquired. Results showed that this framework enabled the robot to obtain several meaningful motions that the operator hoped it would acquire.