Ubiquitous Computing Systems

Research Staff

  • Professor Keiichi YASUMOTO

    Keiichi YASUMOTO

  • Associate Professor Hirohiko SUWA

    Associate Professor
    Hirohiko SUWA

  • Affiliate Associate Professor Manato FUJIMOTO

    Affiliate Associate Professor
    Daisuke NOSE

  • Assistant Professor  Yuki MATSUDA

    Affiliate Associate Professor
    Yuki MATSUDA

  • Assistant Professor Tomokazu MATSUI

    Assistant Professor
    Tomokazu MATSUI

Research Areas

In our laboratory, we are advancing research to popularize Cyber-Physical Systems (CPS), which are key technologies for the realization of "Society 5.0" aimed at making our world smarter, across three focus areas. In "Smart Home" area, we are conducting research on daily living activities recognition and prediction, and smart home appliance control using sensors and machine learning within smart home facilities on campus and in general households. For "Smart Life" area, we are engaged in research that measures and enhances the physical and mental health in daily life and sports using smartphone apps and wearable devices. In "Smart Cities" area, we are conducting research on efficient participatory information collection using an edge-type IoT distributed processing platform, Federated Learning, and gamification, curation that adds value by selecting and organizing collected information, and research on privacy protection mechanisms based on differential privacy technology for securely collecting and distributing various data about people. Furthermore, as part of our smart office research aimed at improving productivity, we are studying context recognition during work and research related to behavior change. All these research areas emphasize operating as a system that includes both hardware development and software implementation, spanning three technological domains: "information collection from the real world" using sensors and IoT, "analysis of information" utilizing machine learning and AI technologies, and "application of analysis results" through apps and services.

Fig.1 Lab overview

Smart home

Fig.2 Smart Home

Aiming to realize a future smart home that is comfortable, energy-efficient, and secure, we have built an actual house on campus, complete with a bath, and are conducting various cutting-edge research projects. These include the development of new sensors, daily living activities recognition using sensors, appliance control, and elderly monitoring. Additionally, we are developing a smart home kit that does not require power supply or communication wiring to sensors, aiming for widespread adoption in general households. Currently, we are working on appliance control that aligns with the residents' schedules and emotions by incorporating real-time activity recognition, future activity prediction, and estimation of residents' emotions and physical conditions. Furthermore, we are proactively challenging new fields by working on micro activity recognition and remote space sharing utilizing 3D point cloud processing technology.

  • Recognition and prediction of daily living activities within the home
  • Elderly monitoring and anomaly detection systems
  • Context-aware smart home services
  • Easy-to-install and user-friendly smart home kits
  • Micro activity recognition and remote space sharing using 3D point clouds

Smart life

Smart life

We are conducting research aimed at recognizing the physical and mental states and their changes associated with various activities (such as daily living, office work, and sports) performed by people, and promoting behavior change that improves the Quality of Life (QoL). In addition to smartphones and wearable devices, we are developing tailor-made IoT (Internet of Things) that embeds sensors and actuators in everyday items, such as chopsticks that recognize eating behavior, chairs that recognize posture, and shinai (bamboo swords) that recognize striking actions. Furthermore, we are researching IoT nudging technologies, including the use of chopstick-type IoT devices that reflect in real-time the type of food being eaten and the speed of eating on an IoT painting (resulting in limited colors for unbalanced diets or murky colors for fast eating), to induce healthier eating behaviors (eating slowly and evenly). We are also working on research into nudging systems that aim to prevent diabetes by predicting the rise in blood glucose levels after consuming a meal and presenting and guiding an appropriate amount of food to prevent postprandial hyperglycemia.

  • Sport/exercise/workout sensing and coaching
  • Estimation of physiological and mental health status
  • Smart office for improving work-engagement
  • Human-computer interaction based on Tailor-made IoT
  • IoT nudge for health behavior change

Smart city

 Smart city

To realize a sustainable society, we are researching technologies for sensing, analyzing, and providing feedback on the state of cities and societies. In particular, we are focusing on participatory sensing technologies that request sensing from people, tourism curation technologies that provide personalized tourism videos, social sensing that extracts valuable insights from social media, congestion sensing using BLE, and estimating human flows through the assimilation of measured data and simulation data. Within these efforts, we are broadly researching technologies necessary for the future society, including mechanisms for obtaining continuous cooperation (such as nudging and gamification), distributed processing mechanisms for sensor data (such as edge computing and federated learning), and sensing and communication methods during disasters.

  • Participatory sensing systems
  • Automatic video curation for smart tourism
  • Congestion sensing
  • Estimating human flows through the assimilation of measured and simulated data
  • IoT platform based on on edge and fog computing
  • Distributed federated learning for tourism object and scene recognition
  • Sensing and communication methods in case of disasters