Masafumi Uchida

Analysis of 1/f fluctuation of keystroke dynamics and heart rate variability

Mental stress has been an issue in modern society. Several studies have investigated the detection of mental stress using physiological indices. However, measurement of these indices would itself impart a certain level of stress to subjects. The temporal dynamics of key typing are called “keystroke dynamics”. Keystroke dynamics show variations with psychological subtlety. The objective of this study was to develop a technique for ambient, unconscious measurement of mental stress based on keystroke dynamics. The relationship between fluctuations in keystroke dynamics and cardiac activity, in the form of heart rate variability, was analyzed. Intervals between key strokes were measured while using a computer under conditions of auditory stimulus or cognitive stress was determined. Power exponent β of the regression curve of the power spectrum for the time series of inter-stroke time (IST), and the power spectrum density of the high-frequency component (HF) for the time series of the R-R interval extracted from electrocardiography were calculated. As a result, the power spectrum of IST approached 1/f with auditory stimulus or cognitive stress, and a significant correlation existed between β and HF.

Quantitative evaluation of body-sway caused by tactile apparent movement

The recognition of tactile apparent movement is normally a subjective sense. However, when applying tactile apparent movement to an engineering system, quantitative evaluation is necessary. In a previous study, we examined the body-sway caused by tactile apparent movement under fixed experimental conditions, but the body-sway characteristics were not fully investigated. In this study, we investigated the body-sway caused by tactile apparent movement under fixed experimental conditions. We focused on biological information and body-sway, and compared cases where the apparent motion was recognized and not recognized. Our findings will help to improve the performance of systems that use tactile apparent movement.

Group behavior of agents with an emotional model

Recently, there has been much interest in the study of the formation of groups of agents that cause interactions between agents and invent new functions. We gave some agents an action rule based on the interactions of human feelings by using a circumplex model. It had been decided that the parameters of feelings in this model should have only two axes. In this report, eight basic action dimensions and pure feelings on four corresponding axes were given to agents as a model of feelings and actions based on the multiple factor analysis theory of R. Plutchik, and the behavioral characteristics of the group of agents were examined.

Autonomous reconfiguration of robot shape by using Q-learning

A modular robot can be built with a shape and function that matches the working environment. We developed a four-arm modular robot system which can be configured in a planar structure. A learning mechanism is incorporated in each module constituting the robot. We aim to control the overall shape of the robot by an accumulation of the autonomous actions resulting from the individual learning functions. Considering that the overall shape of a modular robot depends on the learning conditions in each module, this control method can be treated as a dispersion control learning method. The learning object is cooperative motion between adjacent modules. The learning process proceeds based on Q-learning by trial and error. We confirmed the effectiveness of the proposed technique by computer simulation.

System of personal identification by using tactile stimuli

This paper proposes a personal identification system that uses tactile stimuli. Recently, scientists have been conducting research in biometrics using biomedical information. However, there is the problem that biomedical information is unchangeable. For example, if a thief steals another user’s biomedical information, that person has no information to register. Due to the problems of sight information and unchangeable biomedical information, we propose a solution that uses the sense of tactile stimuli. Tactile information is difficult to steal and relies on human memory, which is unchangeable. This paper proposes a system that uses a pattern formed by a tactile stimuli time series instead of a password number as the identification key. We also discuss the results of an identification experiment and a memory property experiment.

Non-contact measurement of respiratory cycles using a microphone array

Previously proposed methods of respiration measurement have relied on thermistors or strain gauges, which must contact the body throughout the measurement, causing potential discomfort. Here, we develop a non-contact system that measures respiratory cycles by detecting the exhalation sounds using an array of microphones. Several respiratory-sound signals were recorded by the microphone array, and the respiratory signal was separated from the raw measurement data by the independent component analysis method. Respiratory analysis by the proposed method was compared with that of the previously validated thermistor method. Similar measurements were obtained by both methods, confirming that the proposed method eases the discomfort of users without compromising the measurement performance.

Detrended fluctuation analysis of repetitive handwriting

Although there is a tendency that the opportunity to write by hand is decreasing, handwriting is still one of the everyday activities of our lives. In this study, the experimental system that limits the writing speed of a subject was set up. The aim was to measure the fluctuation time series of a time element obtained from the measurement performed using detrended fluctuation analysis. The result confirmed the local self-similarity and that it covers multiple timescales.

Single Trial Analysis of Body Sway Caused by Several Matrix-Shaped Tactile Stimuli on Body Trunk

Postural control is an essential function of the human body. Basically, postural control uses several senses and information from the body as input. After processing the input, the body provides output to the muscles for control. In this report, we examine a method wherein the body is caused to sway in a specific direction using several matrix-shaped tactile stimuli (MSTS) that are located on the skin of the trunk. To clarify the relationship between body sway and the MSTS, we measure the sway caused by the MSTS using a high-speed camera, a stabilometer, and two acceleration sensors. However, each subject’s data exhibits individual differences, and therefore, each trial data exhibits a different feature. We discuss the effects of each MSTS on body sway and evaluate the dynamics of body sway. We then propose a single trial analysis method based on the clustering method and a mixtures of Gaussian.

Propulsion modeling of caudal fin driving system on balloon fish robot

Recently, biomimetic robots combining the characteristics of fish robots and airship robots have been studied. In this work, we consider balloon fish robots (BFR). This paper presents our understanding of the BFR motion obtained through simulations and experimental results. The simulation of the BFR motion is based on a traveling-wave equation, defined as the product of a sine wave and quadratic curve. In this study, we derived an equation of motion for BFR and conducted an experiment to measure the thrust force. By solving the equation of motion with the Runge–Kutta method, we are able to calculate theoretical values for the propulsion velocity of BFR. We validate the simulation by comparing theoretical and experimental propulsion speed values. As a future task, we will measure BFR thrust at more parameters.

Control of body sway using a tactile device

The recognition of tactile apparent movement (AM) is normally a subjective sense. However, quantitative evaluation is necessary when applying tactile AM, an engineering system. In a previous study, we examined the body sway caused by tactile AM under fixed experimental conditions; however, the body-sway characteristics were not fully investigated. In this study, we investigate the body sway caused by tactile AM under fixed experimental conditions. We focused on biological information and body-sway movement, and compared cases where the apparent motion was recognized and not recognized. We propose the possibility of controlling body sway using tactile AM. Our findings will help improve the performance of systems that use tactile AM.