Most humanoid robots lack of dynamic operations such as jumping and running motions. We aim to achieve development of a robot arm mechanism that can throw and pass heavy objects. We had developed a three linked compliant joints robot that has a mechanical adjustment mechanism of joint stiffness at each joint by using a non-linear spring element named as SAT (Stiffness Adjustable Tendon). SAT has a serious problem that the deterioration in the physical characteristics of SAT when significantly large tension is applied to SAT in long period of time. In order to prevent this problem, we had proposed a mechanism named as the Compliance Limiter that limits forcibly a stretching length of the SAT. Four types of three modes SAT with three steps of spring characteristics were developed. As a compliance limiter, a wire cable passes through the center of the SAT using an elastic cord. Because the wire of the compliance limiters is set inside the SAT, the structure is more compact.

抄録全体を表示

Multi-DOF manipulators consist of many actuators. Then, it is difficult to increase output power without they becomes heavier. As a measure for this problem, several mechanisms called "Weight compensation system" have been developed. But these mechanisms cannot be installed in the manipulators which have the roll axis located between two pitch axes. The reason is that there is no reference point of gravity direction in such linkage. In this paper, we propose the new weight compensation mechanism used parallel motion mechanism with bevel gears. It has reference point of gravity direction that is independent from rotation around roll axis, and it is able to be installed the roll axis between two pitch axes.

抄録全体を表示

To avoid collapsing roof due to heavy weight of snow, it is requested to develop a robot that can continuously remove snow from the roof. We propose the Direction-Dependent-Snow-Wiper with the mechanism where the wiper changes its height automatically according to how much snow is. The key idea is that the effect can be achieved without any actuator and passive joint. Through the experimental results by utilizing the scale down model, we could confirm the effectiveness of the basic idea of the proposed mechanism. This paper also discusses the design orientation of the mechanism by introducing the stiffness matrix of the wiper.

抄録全体を表示

Scraping is a very important process in precision machining because not only the quality of the product relate to life. However, because scraping is doing work manually by the expert, there is a problem, such as production costs and the lack of successors. It is that the one way to solve these problems developed a robot that automates the task. In this study, we aimed to reveal the specification in order to realize the scraping operation by the robot, a model is derived from the manual scraping work of a human is determined in the robot.

抄録全体を表示

It is difficult to recognize coil springs randomly placed in a pile by conventional machine vision because of their shape characteristics. In this paper, we propose a method of recognition and pose estimation of coil springs. This method uses their highlights for recognition and the highlights are also used for pose estimation by stereo vision. We construct a bin-picking system using this method and demonstrate successful bin-picking of coil springs.

抄録全体を表示

The purpose of our research group study is to develop an autonomous dual-arm robot for the complex work which involve autonomous cooperation of two arms. The objective of this work is to realize the work to throw a blue ball into a red goal. The coordinate of the ball and the goal is determined by the 3D camera before the action. After the throwing action, the arm will return to its initial position. The robot does this work from starting to ending without human beings interfereing.

抄録全体を表示

The purpose of the our research group study is to develop an autonomous dual-arm robot for the complex work which involve autonomous cooperation of two arms. The objective of this work is to realize the work to draw the characters with a pen that one hand grips on the clipboard that another hand grasps. We implemented impedance control in each arm of dual-arm robot. And the method of cooperative control between right and left arm is showed. As the results of experiment that dual-arm robot draws a line, the control suggested in this study is proved not to be effective.

抄録全体を表示

We are developing a robot system for powerdistribution line maintenance which moves with task instructions, for example "Insert the bolt". This paper shows the task instruction sheet for "Insert the bolt into the hole on the arm tie" and proposes a new task skill for inserting the bolt. In the proposed skill, the arm gropes the bolt near the hole along a spiral pattern to decrease the force in the axial direction (z direction) of the bolt, and then move the bolt to the axial direction and escape the hole edge by detecting the forces in the x, y directions.

抄録全体を表示

Currently, most food processing operations are mechanized in food production. To prevent contamination, the food processing step may produce washing step. There is an automatic vegetable washer and dehydrator that can be done to clean the vegetables. This product had been problem of the yawing vibration due to the unbalance by eccentric load in the dehydration process. This product is equipped with a fluid balancer mechanism as well as the common washing machine as a measure of the vibration. In this research, we describe results of measuring the vibration of the overall rotary tank in the unbalanced state and of the fluid balancer to reduce vibration.

抄録全体を表示

We are developing a robot system for power distribution line maintenance tasks. We analyzed switching gear installation tasks among the maintenance tasks and described task instruction sheets for the installation tasks with a robot language. We have developed a CG simulator to execute robot commands in task instruction sheets. In this paper, we describe a task instruction sheet for "Gripping arm tie on a tool box", a program function for the task command, and CG simulations. The robot can generated the trajectory and move the robot arms for "Gripping arm tie on a tool box" in CG space.

抄録全体を表示

A seven -link dual-manipulator robot with three-fingered hands is proposed with application of vision and impedance controls to produce human-like skills capability for the manipulation of automatic works. First, a stereo camera fixed on top of the robot provides an overview of position of objects placed on a worktable to move the arm closer to the objects. Two case studies have been assigned; in the first case study, the control system of robot hand based on LQI (Linear quadratic integral) control was designed to operate the hand for autonomous work, while the second case, a task of opening a beer bottle was designed to apply this method in the position based impedance control.

抄録全体を表示

Gait rehabilitation is conducted using treadmills or parallel bars. However, there are reports of those who can walk in a training room of a medical facility but cannot in other places. These are because real-life situations totally differ from gait situations in the training rooms in the presence of uneven ground and others cutting across in front of them, so gait rehabilitation is expected to be achieved based on daily living environments for each subject. In this paper, we suggest a walking rehabilitation system for every user.

抄録全体を表示

In the teleoperation system which can provide force display for an operator, the impedance control can enhance maneuverability. However, several sets of impedances decrease force sensitivity of an operator. It is difficult to avoid decreasing it, for there is no quantitative index about it in consideration of mechanical impedances at the moment. In this paper, we propose an index which is expressed with a formula about force sensitivity of the teleoperation system in consideration of them. This formula are derived under the assumptions that Stevens's low are satisfied, and we evaluate to what degree the tendency of the change between the value of it and actual human force sensitivity correspond . From the experimental results, we confirm the validity of it as an index of force sensitivity.

抄録全体を表示

There is currently much discussion of Quality of Service(QoS) measurement at the network level of real-time multimedia services. The presence of a network brings up a number of issues. In case of visual or auditory interaction, the effect of such issues have been treated by many researchers and the effective countermeasure have been proposed. A haptic Interface is a useful device in virtual reality technology and users can make a feeling of touching virtual object. Especially, in a virtual environment connected by a network system, users can share tasks and feelings. However, with a networked virtual environment, the haptic interface will make an unstable behavior because of a communication delay. This experiment focuses on a keyboard interface which uses haptic interacting with virtual space surrounding user. We have also constructed a 3D haptic interaction system, whereby the user can see and touch virtual objects by using a haptic device. However, new methods are known about the relationship between network issues and networked haptic interface in virtual environments. This experiment shows ethernet data transmission on purpose of packet delay and packet loss for transmission, with a few statistical analysis and traditional coding scheme.

抄録全体を表示

This paper describes electro tactile display with micro-needle electrodes that stimulate tactile receptors using electric stimulation. Electro tactile displays can be thin and flexible, however, the high electric impedance of the stratum corneum requires high voltage to stimulate the tactile receptors. In our device, we use needles as electrodes and the needle length was designed to have approximately the same length as the stratum corneum so as not to injure nerve area. This device allowed subjects to feel tactile sensation with lower voltage than a flat electrode array with good repeatability. In addition, the tactile sensation was created in the vicinity of the needles, i.e., tactile display with a high spatial resolution was achieved.

抄録全体を表示

This paper reports development of MEMS displacement amplification mechanism using latex. The actuator which was developed by our laboratory consists of piezoelectric actuators and hydraulic displacement amplification mechanism (HDAM) that encapsulates liquids in micro-chambers with two flexible polymer membranes. HDAM can amplify the displacement of MEMS actuator, so the actuator can output tactile feelings to human's fingertip. In order to increase amplification ratio of HDAM and give textures was created by tactile display more smooth, we developed new construct and fabrication process of HDAM. We used the latex for material of membrane was used HDAM contact part. For making more smooth textures, we developed the tactile display which has 5 x 5 large-displacement MEMS actuators. At last, we investigate the characteristic of new HDAM and successfully showed the potential of creation better tactile feeling by great amplification ratio and drive frequency because of using latex.

抄録全体を表示

In the manufacturing industry, the shortage of skilled workers has been pointed out due to the aging problem. In order to solve it, an interactive work support has been studied. It is a technology that a worker can receive appropriate working instructions according to the situation at that time. To realize the technology, we focused on display technology for presenting the work instructions. And a video see-through display that can present images with full HD has developed. When an operator performs a detailed task looking at the working object through such a display system, it is necessary to obtain a sufficient depth perception. In this paper, some evaluation results of depth perception using high-resolution binocular video see-through display that we developed are reported.

抄録全体を表示

The purpose of this research is to develop an interface which enables us to operate TV's by hand motions. We use a stereo camera to recognize a hand motion. A fast image processing is necessary to recognize quick hand motions. So we always track a person in front of a TV by using upper body in depth images and face, and process only an area surrounding him/her. The fingertip position is detected as the most protruded region in the depth image around the faces. In gesture recognition, the trajectory of the fingertip are analyzed to classify the hand motion into a predetermined set of gestures. In pointing recognition, we calculate the corresponding between the position of the fingertip and that on the display. The relationship between the coordinate system in the image and that on the display enables to mapping the fingertip position onto the display.

抄録全体を表示

In this study, a new method is proposed to enhance operational performance with surgical robot using the relation between haptic perception and human myoelectricity. Master-Slave type of surgical robot is utilized in this research. The robot is able to display force acted on the target by using force feedback. We experimentally distinguished operators' types of haptic perception by tests under relaxed and strained operational conditions. Then we measured EMG value under these conditions and got the EMG characteristics of certain operators. An empirical formula of force scaling factor was found to indicate the scaling factor with EMG value. Force feedback to the operator is altered by the scaling factor. Better operational performance was achieved by modifying scaling factor according to the haptic perception of the operator.

抄録全体を表示

In recent years, virtual reality research is treated a virtual object mainly as the elastic body and solid. However, the object in the reality world has many not only the elastic body and solid but also things including the fluid. In addition, fluid simulators have many images without sound although fluid simulation has been successful. We know that a sound is necessary from experience because we live while being surrounded by the world with the sound. Therefore we can expect expandability of virtual reality space by adding a sound to simulation. This research, it aims at adding the fluid sound in the real-time fluid simulation. As the first stage, we pay attention to air bubbles sound and implement the simulation. We tried the determination method of reproduction sound based on a physical phenomenon and the sound generation by artificial sound.

抄録全体を表示

One of the medical issues of our country is increase of the burden of doctors per capita by aging society and localization of the physician. Diagnosis and treatment support system is necessary to reduce the doctor's load. In this paper, we have experimented for analyzing an ultrasound probe scanning approach by the measurement interface. First, we decide definition of skillful procedure for echography. And, we decided target of the basic cross-sectional imaging. Next, we have developed the calibration method on body surface. Finally, we did experiment with measuring the ultrasound probe approach on several patterns. As a result, we got two skillful physician's knacks. One is start position of probe approach. Other is probe moving method of short axis from long axis. We succeed in clarify knack of probe approach by skillful man.

抄録全体を表示

One of the medical issues of our country is increase of the burden of doctors per capita by aging society and localization of the physician. Diagnosis and treatment support system is necessary to reduce the doctor's load. In this paper, we have constructed trajectory measurement system of ultrasound probe scanning for analysis of procedure in echography. First, we constructed measurement method changed to DLT method by GigE cameras from ARToolKit by web cameras. We conducted an experiment measuring the position and attitude accuracy. Next, we confirmed that when measuring the trajectory scanning, the resolution capability is improved compared to previous systems. Based on the above, we have experimented scanning probe measurements. And we succeeded in visualization of measurement for ultrasound probe scanning of skillful and unskilled. In addition, the occlusion problem that multiple cameras lose sight of the marker could be resolved by increasing the number of cameras.

抄録全体を表示

We describe a haptic device which is aimed at presenting the sense that a forefinger traces an object and which is made up of nonmagnetic metal and a permanent magnet. This paper shows we make the experimental model which is composed of a high speed camera and automatic stages and which changes the produced power by controlling the velocity of automatic stages which are mounted magnets. In addition, we measure the generated power with a load cell in order to verify that an experimental model can generate force in accordance with theory.

抄録全体を表示

Nowadays, controlling vehicles through video image-based interfaces is becoming more common. Tele-robot operation is one of applications of those interfaces. Thanks to advanced functions of driving assistance systems, it is possible to automatically avoid actually crashing with obstacles. However future prediction in real time route planning during actual operation is a challenging issue for operators. This paper introduces a vehicle operation interface which presents a predicted future. The interface consists of a real time depth image based on modeling technique of real environments and a trajectory prediction of a vehicle. The predicted vehicle interacts with the real world by using a physics engine on Augmented Reality. It assists human ability to control vehicles in unknown real environments.

抄録全体を表示

In Japan, the population of person requiring support is increasing. Exercise is important for care prevention. So, local government has carried out gymnastics for care prevention. However its participation rate is low. Therefore, improvement of the motivation to exercise is needed for elderly people. So, we develop input device which operation will be exercise for the preventive care and combine it with tele-robot. Then tele-robotics contributes to daily life and hobby of the elderly. And we expect the elderly can improve the motivation to exercise. In this study, we improved device developed before and conducted the simulation in order to evaluate the operativity of the device.

抄録全体を表示

This paper addresses a gravity based haptic device that does not need to be flexed to ground. The device gives rotational moment in hand to user by controlling its center of gravity. Furthermore, it can be used as a steering controller by gyro sensor implemented on the device. Finally, two basic experiments are carried out for verifying the controlling performance with tactile support.

抄録全体を表示

Cuddly is a mobile phone application which will enchant soft objects to enhance human's interaction with the objects. Cuddly utilizes the mobile phone's camera and flash light (LED) to detect the surrounding brightness value captured by the camera. When one integrate Cuddly with a soft object and compresses the object, the brightness level captured by the camera will decrease. Utilizing the measurement change in brightness values, we can implement diverse entertainment applications using the different functions a mobile phone is embedded with, such as animation, sounds, light etc.

抄録全体を表示

A Head Mounted Display (HMD) is the most popular display device for Virtual Reality (VR), Augmented Reality (AR), Mixed Reality (MR), Telexistence, and remote Behavior Navigation. However, it is difficult to realize both large field of view (FOV) and natural correspondence between the displayed image and the real image with an HMD. In order to improve the performance of the HMD, we propose the simultaneous utilization of both a slenderized HMD for central vision and a projection display for peripheral vision. In this paper, the configuration of the prototype of the HMD/projector hybrid display system, and the experimental evaluation are presented.

抄録全体を表示

This paper describes an annotation system using motion capture data to teach naginata performance. There are some video annotation tools such as YouTube. However these video based tools have only single angle of view. Our approach that uses motion captured data allows us to view any angle. A trainer can write annotations related to parts of body. We have made a comparison of effectiveness between the annotation tool of YouTube and the proposed system. The experimental result showed that our system triggered more annotations than the annotation tool of YouTube.

抄録全体を表示

We developed a system that provides a human user with the artificial oculomotor ability to control their eyes independently, which was named Virtual Chameleon after Chameleons. The successful users of the system were able to actively control visual axes by manipulating 3D sensors held by their both hands, to watch independent view fields presented to the left and right eyes, and to look around as chameleons do. Although it is thought that the movements of camera angle of Virtual Chameleon corresponds to smooth movements on human eye movements, the system did not have functions to control angles of cameras corresponding to saccadic eye movements. We implemented functions to control angles camera rapidly corresponding to saccadic eye movements in Virtual Chameleon. In this paper, we performed fundamental experiments to evaluate effects on the user's visual capabilities to investigate human performance.

抄録全体を表示

We scanned 3D shapes of urban area by using a laser range sensor to build a 3D map for mobility robots. Different from usual 3D maps such as Google Earth and commercial navigation systems which focus on buildings, our map focuses on lower height parts. Since our aim is to build a specialized map for robots to move automatically in urban scene, further details are necessary such as hedges, guardrails, differences in ground heights. In order to digitize the Tsukuba mobility robot special district, we connected many scan data sets to align the unified coordinate system, and filled range data blanks. In addition, we discuss the future prospects with respect to the 3D map for mobility robots.

抄録全体を表示

In order to support the short-distance transportation of the elderly or those with walking difficulties in the city, we are developing a single-passenger mobility-support robot "ROPITS", which runs on pedestrian road autonomously. In this paper, we introduce the ability to run the robot to any point specified in the map of the portable information terminal. Since the robot estimates the latitude and longitude of its location with high accuracy, it is capable of traveling to the point specified with a general map. We report the experimental result of the robot running in the Mobility Robot Experiment Special District in Tsukuba-city, where mobility robots can run on the public pedestrian road.

抄録全体を表示

In this paper, we propose a data structure of 3D map for sharing between many mobile robots via cloud computing. To realize autonomous traveling in area that a robot never visited, we consider the robot should reuse maps that are constructed by other robots. For the sharing, the data structure of maps should become efficiently retrievable against searching queries. This paper presents a data structure for 3D maps by using some database techniques those are Hash Map and UB-Tree. By using Hash Map, the proposed structure can achieved fast data access. By using UB-tree the proposed structure can achieved minimum memory usage.

抄録全体を表示

Tsukuba city office supports the creation of robot industry and its verification in real world, and aim for a society that achieves a more affluent and comfortable daily life by using robot technology. In 2011 Tsukuba was approved as first mobility robot experiment special zone which enabled the boarding type robots to travel on public roads. This paper explains about activities in mobility robot experiment special zone

抄録全体を表示

In order to entry the Special zone of experiment by mobility robots, it is required to obey the inspection of District Land Transport Bureaus in development of mobility robots. This report describes a challenge of development for a personal mobility robot that makes fun to ride. The mobility robot has two front driving and one rear free wheels, and the center of rider's body is placed on upper of the rear wheel. Therefore, the rider has fun since feels like rotating its body when the robot moves on curve. Furthermore, for fun, the shape of the robot is designed to be stylish since it is considered that its looks are important. As the results of developing the robot, this report shows the basic design for the inspection, assemble the chassis by using general parts, and molding the exterior body for stylish and safety.

抄録全体を表示

In this paper, we describe a position determination method using the Reinforcement Learning in the RoboCup soccer. In the RoboCup Soccer, one of the aims is a realization of cooperative action among multiple robots. In previous studies, a localization algorithm in dominant area has been proposed. However, it is difficult that multiple robots behave adaptively using existing methods in a dynamic environment like the RoboCup Soccer. In order to solve the maladjustment of such an environment, we propose a method for position learning method using the dlaunay triangulation and the reinforcement learning,. The proposed method is applied to the RoboCup soccor simulation and usefulness of the method is confirmed.

抄録全体を表示

Recently, a lot of robots are used for industrial, entertainment, research and so on. For example, in entertainment robot, RoboCup is one of the most active robot competitions in the world. RoboCup main topic is autonomous soccer robot competition. A RoboCup soccer robot recognizes the object by color. Therefore, if the RoboCup soccer robot want to be made for intelligent, robot needed very good image processing system at first. In this research, we aim to participate in the RoboCup Middle Size League competition. In this paper, we report about how to develop an image processing system using the color information.

抄録全体を表示

Robot industry is expected to grow because address the needs of a low birthrate and an aging population. In recent years, development of indoor robots becomes popular and RoboCup@Home which is the at-home robot competition is held every year."RoboCup@Home league" works on various problems in the home environment and teams have to compete to solve some tasks. Through this competition of an autonomous robot designed to support human being in daily life, We aim at the realization of a practical robot doing even more useful work. This paper shows Omnidirectional Wheel Drive Robot for RoboCup@Home and also shows the system as well as future prospect.

抄録全体を表示

RoboCup is an international join project to promote Artificial Intelligent (AI), robotics, and related field. It is an attempt to foster AI and Intelligent robotics research by providing standard problems where a wide variety of technologies can be integrated and examined. In RoboCup, soccer game is selected as a main topic of research, aiming at innovations to be applied for socially significant problems and industries in the future. We also have organized RoboCup team "Hibikino-Musashi" Middle-Size League soccer competition. In this paper, a new fast moving mobile platform "Musashi150", which moves maximum 3.5 m/s, is introduced.

抄録全体を表示

We introduce developed autonomous soccer humanoid robots for RoboCup2012. We started to develop the robot from 2006. Now, we are developing the forth generation humanoid robot for the kid size league. Through these developments, we studied on the mechanics, electronics, system design, control algorism, educational application and so on. We sometimes reported the problems of the systems and proposed the methods to solve these problems. In this paper, we report the RoboCup2012 humanoid league and the result of our robot system. And, we introduce the forth generation humanoid robot "Dynamo Accelite". The robot was designed to solve some problems of the previous robot "Dynamo".

抄録全体を表示

We are studying on the influence of the state transition of touch ground on the biped robot walking. We have been developed a small humanoid robot "Dynamo" for the RoboCup. "Dynamo" can walk stable and fast relatively. However, in the RoboCup2010 Singapore, the robot can not walk stably because the field is slippery. Then, we found the robot can walk stably if the sole angle is changed in trial and error. In this paper, we analyze the influence of the sole angle and discuss why the angle is important for the stable walking. Then, we propose a walking method and confirm its effectiveness through the experiments with the humanoid robot "Dynamo".

抄録全体を表示

This paper presents a novel computational approach for modeling and generating object manipulation behaviors by a humanoid robot. A time-delay deep neural network is applied for modeling the multiple behavior patterns represented with multi-dimensional visuomotor temporal sequences. Thanks to the efficient optimization capability of the Hessian-free deep learning algorithm, the proposed mechanism is successfully trained with six different behavior patterns without any dedicated feature extraction preprocessing. The experimental results show that the proposed mechanism can perform as a temporal sequence predictor as well as a cross-modal memory retriever. The result from a robot control experiment in the real environment demonstrates that the proposed mechanism can be utilized for adaptively selecting behaviors in accordance with the consecutive environmental changes.

抄録全体を表示

Recently, a variety of researches and developments on BMIs attract many researchers' attentions. However, most of these BMI researches intend to assist the disabled. The purpose of this study is the development of real-time power assist technology using BMI that can also be used even for healthy people. To achieve this we consider that motion identification by EEG is necessary. In this paper, we analysed the experimental data and tried to clarify the relationship between EEG and load by wavelet transform. From the results, some interesting characterristics are obtained. It shows the possibility to extract the motion information from EEG.

抄録全体を表示

This paper describes a novel dynamical neural network model for learning and generating object manipulation behavior. The network learns to predict not only the mean of the next input state, but also its timedependent variance. The training method is based on maximum likelihood estimation by using the gradient descent method, and the likelihood function is expressed as a function of the estimated variance. Regarding the model evaluation, it was shown that a humanoid robot with the proposed network can learn multiple behavior of object manipulation and adaptively select trained behavior in accordance with the environmental changes.

抄録全体を表示

Recently, Brain-Machine Interface (BMI) system is focused in robotics and medical science. In this research, we aim to design a BMI system using Near-InfraRed Spectroscopy (NIRS) which measures brain blood flow change. This time, we discuss a method of classifying thought with potable NIRS system in the living environment. In this paper, we discuss a method to increase classification performance in the living environment.

抄録全体を表示

In this study, we built an infant size robot with a vision system and a musculoskeletal system in order to investigate the effect of the interaction between a sensory system, a musculoskeletal system, and an environment on development of crawling motion. We found that the robot could learn a crawling motion by optimizing parameters of neural oscillators, using a reward function based on subjective visual information. We also found that a distance from the robot to a target object makes a difference in the convergence time of the learning process due to utilizing the subjective visual information.

抄録全体を表示

We analyzed properties of a brain responding to unnatural sensory feedback by using a brain-machine hybrid system. The brain-machine hybrid system is a two-wheeled mobile robot with an insect's head as a controller. On the robot, we recorded neural signals relating to steering behavior and translated them into angular velocity of the robot. Because the hybrid system has sensors on an insect, we can give disturbances to control sensory feedback by manipulating motor output of the robot. By calculating neural activities responding to disturbances, we could observe that an insect cancelled out rotational disturbances. Moreover an insect kept motor output of the hybrid system by modifying neural activities according to conversion rule during odor source localization behavior.

抄録全体を表示

This paper proposes methods to generate sentences from motions by using a large word dataset, and to distribute the computations to cloud computers. These methods make it possible for humanoid robots to understand human motions in various sentences in real time. The proposed framework consists of two modules for associations between motions and words, and sentence structures of word trigrams. The framework recognizes motions, associates words with them, and arranges the words into sentences by taking into account their lengths. We tested the framework on captured motion data and a large scale dataset of word Ngrams by using cloud computer of Amazon Elastic Compute Cloud, and demonstrated its validity.

抄録全体を表示

Brain-signal has been studied as an input modality for human-machine interface (HMI). Using Brain-signal-based HMI, people with impaired abilities can communicate with a machine by the brain's electrical activity. This study focused on Electro Encephalo Gram (EEG) signal measurement and analysis methods related to concentration for multimodal Interface. The experiments have been performed with various tasks. The result of this study showed there are some meaningful results about threshold, selfarithmetic and eye-close activity. It can be used as a modality for brain-achine interface.

抄録全体を表示