Start Global Navigation

  1. Home
  2. About RCAST
  3. Research
  4. Industry-Academia-Government Collaboration
  5. International Collaboration
Research

Start Main Contents

Building a future with Big data
Hiroyuki Morikawa (Networks and Mobile Systems)


  "Big Data" is the most popular new key word in the IT field today. Hopes are high for the potential to use big data to create new value in various fields. The Morikawa Laboratory (led by professor Hiroyuki Morikawa) is developing core big data technologies, as well as engaging in applied research in using big data to create innovative services, contribute to a safer and more secure society, improve industrial efficiency, and more. How will society change as the result of big data? We visited this cutting edge research laboratory, creating a new information society, to find out.


Morikawa Lab
The Morikawa Laboratory's "ROSSO System".
This area is used to provide demonstrations to visiting high school students, companies, and ministry officials.







PASERI
Verification testing of sensors for visualizing crop growing conditions. Development is also underway on a "Smart Greenhouse" which automatically controls the growing environment.





■Big Data's Potential

  RCAST Building 3 is home to the Morikawa Laboratory's verification testing space, "ROSSO System".
Big data-related verification testing and demonstrations are held in this stylish site. It doesn't have any large experimental equipment, but it is designed to allow flexible installation of sensors and actuators, and also serves as a living room for researchers. Professor Morikawa explains, "By conducting experiments in a living room, we are able to perform verification testing of big data applications that are tightly focused on people's daily lives."
  Big data, as its name indicates, refers to massive collections of data. Not only is the amount of data large, but the data is also extremely diverse. Even as you read this, new data is constantly being created in real-time.
Professor Morikawa divides big data into two categories. The first is virtual data, generated by Internet-connected computers. This includes social networking data and online shopping purchase history data. The other is real data, produced by sensors and devices. Particular attention is being focused on this machine-to-machine (M2M) big data. "M2M big data has unlimited potential for improving productivity in areas such as agriculture, medical care, and distribution," says Professor Morikawa. That is because this M2M big data is "information which until now could only be recognized by expert observers."


■Visualization of Crop Growing Conditions

  The Morikawa Laboratory has developed numerous core technologies for collecting M2M data, such as sensor networks and wireless transmission technologies. Since last year 250 sensors have been installed inside RCAST, successfully collecting temperature, humidity, and other data at each sensor position and transmitting the data via the sensors. This technology has made it possible to collect data which was impossible to collect in the past. Merely collecting massive amounts of data, however, is pointless unless that information is then used. How can collected M2M data be put to use? "One possibility is agriculture. At the Morikawa Laboratory, we're now conducting verification experiments in which we visualize the growing conditions of crops," explains Professor Morikawa. Sensors have been installed in a greenhouse to measure weather data such as brightness. These sensors are collecting data on light transmission rates to estimate leaf surface areas. The leaf surface area can be used to predict optimal harvest times. Many farmers currently check fruit and vegetable growing conditions visually. A system which assessed growing conditions based on objective indices could potentially promote entry into the agricultural field by new farmers and lead to efficient agricultural management which doesn't depend on personal experience.


Crop Sensing Platform for Improving Agricultural Efficiency 区切り線

Farming relies on the experience and intuition of farmers. This creates a hurdle to new entry into the agricultural field, and produces inefficiency. Collecting and analyzing data could make farming efficient. This is the foundation of the new "evidence-based farming" agricultural management approach.

Why measure "light"?
Leaf growth is an indicator of crop growth. When plants have many leaves, they block out light, making the area under the leaves dark. Light transmission rate measuring sensors can be used to assess plant growth and help predict optimal harvest times.

Sensing Platform
PASERI:Photosynthetically Active Radiation Sensor for Evidence-based agRIculture

内部リンクNEXT    Solving Social Problems with Big Data   >>

Start Site Information

page top

Copyright (c) RCAST, The University of Tokyo