We conduct research in the field of Assistive Living Engineering, aiming to build a society that is safe, secure, and comfortable by focusing on interfaces in daily life between people and activities.
Interfaces and assistive technologies that promote independent living for older adults and people with disabilities
To improve safety, comfort, and usability in mobility for older adults and people with disabilities, we objectively analyze the characteristics of users’ remaining functional abilities and develop interfaces that reflect these characteristics. In this research, expanding each individual’s “capable functions” is linked to motivation for maintaining health. We further regard the user’s physical functions and self-regulation ability as a cooperative learning system during mobility, and design intelligent mobility systems that can adapt to changes in these characteristics. By repeatedly prototyping the system and conducting field trials, we work toward developing safe cooperative control systems as integrated human–machine systems.
Adaptive intelligent mobility technologies
To achieve autonomous mobility using camera-based visual recognition without relying on high-precision maps, we learn the behaviors and knowledge by which humans adapt to and harmonize with driving environments. By modeling both (i) how people adapt their behavior to diverse driving environments (geometric/structural changes, infrastructure information, interactions with pedestrians, and interactions with other mobility devices) and (ii) how the environment changes in response to those behaviors as probabilistic transitions, we build prediction and recognition systems required for autonomous mobility. Based on these characteristics, we develop autonomous mobility systems through path planning and safety assurance functions that prevent collisions with obstacles. We validate the developed systems through prototyping and real-world evaluations, and apply them to autonomous wheelchairs in large facilities such as hospitals, airports, and shopping malls, as well as delivery robots and last-mile mobility in rural areas, including unpaved roads. In addition, to enable low-cost, safe, and reliable mobility, we are advancing this research by incorporating data-driven functions as a cyber-physical system, such as infrastructure cooperation and cloud integration.
Extraction of behavioral features that change with health and aging, and human–machine cooperative systems based on these features
To establish support strategies for transferring skills from expert operators and for preventing accidents among older drivers, we structure the age-related changes in physical and cognitive functions together with the experience and knowledge that compensate for these changes. By identifying critical variables that lead to high-risk (unsafe) behaviors, we develop support systems for human–machine cooperative settings. Examples include real-time skill acquisition systems for novice operators and systems that suppress unsafe driving behaviors when functional decline is detected. The outcomes of this research contribute to skill transfer from an aging population of expert operators. In addition, by comparing behavioral characteristics of healthy older drivers with their mind–body functions, including brain function, this work also supports medical–engineering collaboration toward exploring new clinical insights and potential pathology discovery.
Comfort-oriented product design using physiological and behavioral information
In the research area of usability when people use products and systems, we propose objective evaluation indices and build systems to assess users’ mind–body functions and their changes over time. Our goal is to extract key factors that contribute to comfort in the design of products and systems. We view comfort-oriented design from two complementary perspectives: “pleasantness,” which evaluates user sensibility based on state, experience, and knowledge, and “fit,” which anatomically understands and evaluates human functions. By quantifying physical workload during behavior and operation as well as physiological signals, we develop systems that support product design and systems that enable comfortable autonomous mobility. These methods have been applied to practical product development—including vehicle bodies and seats, bathroom and toilet spaces, PET beverage bottles, and cosmetic tools—and have led to commercialization.