Koji Yamada

Haptic Communication Robot for Urgent Notification of Hearing-Impaired People

A proposed robot provides functionality inspired by the behavior of hearing dogs, which assist hearing-impaired people by alerting them to important sounds such as fire and smoke alarms. Hearing dogs use body contact for communication with hearing-impaired people, a process heavily depending on haptic communication. The proposed robot also uses physical contact for urgent notification, like a hearing dogs touching behavior. The robot can move using its wheels and then approach and bump people. To be explicit our proposal, our bumping behavior model was simplified and an experiment was conducted with hearing-impaired participants and a prototype robot. This paper reports the experimental results and discusses effectiveness of the robots bumping behavior.

A supporting system for colored knitting design

Knitting needs some artistic sense and technical knowledge. We had already proposed the supporting system which changes a simple design figure which knitting designers construct into a pattern-knitting diagram by changing rules. Generally you use various colored strings to knit. However, the system had dealt with only monochromatic knitting patterns by now, and a kind of system representing colored knitting patterns could support many knitting designers. Therefore in this paper, a method generating colored images of complete knitting pattern is described.

Edge Extraction Using Fuzzy Reasoning

We characterize the problem of detecting edges in images as a fuzzy reasoning problem. The edge detection problem is divided into three stages: filtering, detection, and tracing. Images are filtered by applying fuzzy reasoning based on local pixel characteristics to control the degree of Gaussian smoothing. Filtered images are then subjected to a simple edge detection algorithm which evaluates the edge fuzzy membership value for each pixel, based on local image characteristics. Finally, pixels having high edge membership are traced and assembled into structures, again using fuzzy reasoning to guide the tracing process. The filtering, detection, and tracing algorithms are tested on several test images. Comparison is made with a standard edge detection technique.

Behavior Model for Hearing-Dog Robot

Hearing dogs play a role of assisting hearing-impaired people. The touch of the dog to the people can alert the people to important life sounds, such as doorbells, alarm clocks and fire alarms. These dogs are professionally trained to touch and lead the handler to the source of the sound. Meanwhile hearing dogs are insufficient in the world. Thus a hearing-dog robot is expected as an alternative assistant. In this paper we proposed and constructed a behavior model for the hearing-dog robot. The behavior model is constructed by a discrete state transition. The robot can change the state and act properly according to the information acquired from the sensors of the robot. We conducted an experiment to evaluate the proposed behavior model. In the experiment the robot actually could act according to the behavior model, but in some cases the robot failed to notify the target user. We discussed the causes the robot failed and the improvements to solve them.

Touch-based information transfer from a robot modeled on the hearing dog

Research on physical human-robot interaction has been attracting attention recently, focusing on robot embodiment. The work reported here proposes Active Touch Communication Robot (AcToR), a robot that is modeled on the hearing dog. A hearing dog is a type of dog assist people who are deaf or hard of hearing by alerting their handler to important sounds. AcToR uses the sense of touch to notify a human of the intention to transfer information. For example, when AcToR detects that a cell phone that is in another location has received a call, AcToR moves to the users location and makes contact with the users body to notify the user of the incoming call. The AcToR robot is based on the Roomba® and uses the Roombas bumper and contact sensors to detect contact. This paper reports the results of psychological experiments using the AcToR robot that indicate the feasibility of using touch to transfer information from a robot to a person.

Can a Robot Wake a Sleeping Person Up by Giving Him or Her a Nudge

In this study, a robot is modeled to have the behavior of a hearing dog that assists hearing-impaired people. Hearing dogs use physical contact to communicate with hearing-impaired people. The proposed robot nudges an object in contact with a person (in this case, a bed) to deliver urgent notifications. This paper describes the concept of a robot with a nudging capability and discusses an experiment to determine whether such a robot can wake sleeping people wearing earplugs. We discuss the functions of the hearing-dog robot as a wake-up alarm.

Multi-agent cooperative reasoning using common knowledge and implicit knowledge

This paper presents a method for multi-agent cooperative reasoning using common knowledge and implicit knowledge.

Hearing-Dog Robot to Wake People Up Using its Bumping Action

We propose a robot inspired by the behavior of hearing dogs. Hearing dog is a guide dog for deaf people by notifying them of some life sounds such as fire alarms , doorbells, wake-up calls and so on. Hearing dogs use their touching behavior to communicate deaf people. This is a kind of haptic communications. The proposed robot also uses physical contact to communicate the people like hearing dog’s touching behavior. The robot can approach and bump the people to notify of some life sounds happened around the people. This paper utilized the robot as wake-up call. We conducted an experiment to evaluate the usefulness of the robot to wake up sleeping people. This paper reports the experimental results and discusses usefulness of the robot’s wake-up call.

Black-and-white drawing support for adobe illustrator using onomatopoeia

Adobe Illustrator provides many tools for a user to create art works. The tools also provide many parameters or options that the user can tune up or select. This study focuses on black-and-white drawings (“Sumi-e” or “Suiboku-ga” in Japanese) on Adobe Illustrator. In order to create black-and-white drawings, the user usually has to select a proper brush from preset brushes Adobe Illustrator provides. But the selection is difficult for the user, especially beginners. This paper proposes a supporting plug-in tool for the user to decide and apply the proper brush. The plug-in can accept an onomatopoeic utterance input which expresses the users own imagination or impression of the brush. Onomatopoeia is well known as a useful figurative expression for Japanese people to describe their own imagination or impression for something. This paper reports the configuration of the plug-in and illustrated some examples of black-and-white drawings created by using the plug-in. The architecture of the plug-in consists of CNN(convolution neural network), database of brush samples, brush image interpolation, fuzzy-based image expansion and contraction, and so on. CNN allows input of arbitrary onomatopoeias. The brush image interpolation generates a new brush from brush samples in the database. The fuzzy-based image expansion and contraction transforms the new brush into a natural-looked brush form with less jaggy.

Japanese Kanji-calligraphic font design using onomatopoeia utterance

Onomatopoeias are useful tool to communicate intuitively and conveniently among people. This paper focused on the utility of onomatopoeias and proposed a design supporting system of Japanese Kanji-calligraphic “Shodo” font using onomatopoeia utterance. The system can generate a variety of scratched- and blurred-look design of “Shodo”. In order to realize the design, the system utilized the correspondence between onomatopoeias and the scratched- and blurred-look. This paper reported the concept and the architecture of the proposed system. And we demonstrated some output examples generated by the system.