Under the theme of "making robots intelligent," various basic technologies are also being studied. These include the "self-position estimation problem," where the robot ascertains its own current position as it moves; the "autonomous distributed system," where several autonomous robots perform a task through mutual collaboration; "machine learning," where the autonomous robots change their behavioral patterns in accordance with surrounding environment; and the "man-machine interface" that aims to mutually connect humans and machines. The research themes are diverse, and the laboratory is engaged in various challenges in materializing an intelligent robot.

• Read More (in Japanese)

The convenient and smart world of the Internet is causing problems concerning information disparity and information errors. Information devices in the future will need a mechanism to communicate information easier to comprehend, while giving due consideration to a person being communicated to. Our laboratory aims to build robots and devices, whereby the status and latent requirements of the person are grasped by observing conversations and gestures, and to provide information easily comprehensible to anyone by using timing and methods suitable to the situation. To achieve this, research is being conducted on conversation analysis technology using audio and video mechanisms to grasp the status of the person, and technology for natural interaction between humans and machines.

• Read More (in Japanese)

As artificial intelligence and robots become more and more prevalent in society, the question now is, "How will the Internet and computers affect our learning?" Our laboratory is engaged in research on new ways of learning that will be required by society in the future, such as developing teaching materials to improve human learning using information and communications technology, developing learning tools, and measuring the effectiveness of curriculum and learning.

• Read More (in Japanese)

Although high-performance and high-precision robots have been developed successfully, robots that can flexibly and rapidly adapt to changes in the environment in a human-like manner have yet to be developed. Our laboratory aims to develop robots that can move as skillfully as human beings by modeling the control mechanism of the living organisms and applying it to the design of the robots.In particular, our laboratory is engaged in the development of a motor-less power- assist equipment and the skillful adaptive motion control of life support robots.

• Read More (in Japanese)

Our laboratory conducts research with the aim to create a valuable application using human sensing measurements. Sensing targets include not only biological signals such as electrocardiograms and electromyograms, human motion, joint angle, forces that directly measure humans, but also sensing voice and device operations. Currently, our research focuses on the assistance of human activity by using an advanced method of human sensing.

Main Research Topics

• A sit-to-stand assistant system using predict motion technology using EMG
• Monitoring system for the elderly using wearable devices
• Automatic image switching method for smart glasses based on user sensing

• Read More (in Japanese)

Human-friendly robots must act in accordance with human intentions and feelings with ensuring safety and a sense of security for people. We conduct research on stress-free methods of measuring biomedical information of people such as heartbeat, breathing, eye movements, brain waves and myoelectricity and also on technologies for inferring their psychophysiological conditions based on the information. In addition, we develop systems applying such measurement technologies.

• Read More (in Japanese)

With advances in computers, the expressive power of images has become significantly richer. However, such expressive power is specialized only for audiovisual modalities. Moreover, the devices used to input data into computers, such as the touch panel, keyboard, and mouse are not “natural” to humans. In this laboratory, research is being conducted on technology that facilitates natural interaction between humans and computers by using the natural action of humans as input, and providing responses from the computer not only for audiovisual perception but also for the five sensory perceptions, including tactile perception.

• Read More (in Japanese)

Our laboratory perform research on the human phenomenon of "viewing," from landscapes seen from mountaintops to railway station signage and smartphone screens. We propose a new method for evaluating human sensibility by conducting experiments to understand the brain waves and other biological information that are activated when people view landscapes, townscapes, and different types of design. Subsequently, the measurement data acquired from the experiments are combined with the results of behavioral studies to improve their usability in the fields of civil engineering and architectural planning, new-product assessment, and digital signage.

• Read More (in Japanese)

We use intelligent system technology to realize learning, prediction, and optimization for social systems and biological systems.
Our main research topics are:
1. Data recognition and generation: Deep neural network is used for data-driven learning to classify 2D or 3D data. Our research targets include image style transfer, 3D building structure design and 4D phenotype analysis of a biological system.
2. Optimization: Several approaches from mathematical programming to meta-heuristics are used for the optimization. Our research targets extend from VSLI floor planning by simulated annealing, inventory control by demand prediction for optimal logistics, to power management in a decentralized network of renewable energy by stochastic programming.
3. Bioinformatics: Machine learning is also used in bioinformatics. One of the present targets is phenotype analysis using 4 D dynamic data on cell division in early embryo of C. elegans as a joint research with a biologist group.

• Read More (in Japanese)