Our Laboratory

General Overview

Computer networks such as the Internet continue to increase in speed and capacity. This network is the world's largest system used for exchanging information. On the other hand, systems that simulate human muscles and joints are actively developed, as well as highly functional high-performance robots. In this laboratory, we extensively studying the control technology of such logical and physical systems.

About research on network systems

The proliferation of smartphones has led to a dramatic increase in traffic flowing through wireless networks. This has created a problem of limited a spectrum band. In response, new wireless communication technologies such as LTE, WiMAX2, and IEEE802.11ac have been developed and put into practical use. However, these technologies only improve the communication speed of each individual system and do not improve the overall efficiency. Therefore, in our research, we construct heterogeneous wireless networks that integrates and interconnects these systems.

In recent years, there have been growing expectations for new services provided by machine-to-machine (M2M) and Internet of Things (IoT) communications. However, conventional networking technologies cannot be directly applied to these services because each node is characterized by extremely low functionality while being connected to a large number of other nodes. Therefore, we study the construction of wireless multi-hop network technologies to support these M2M/IoT communications.

Achievements are listed here

Studies on systems control

The mathematical model of an object is indispensable for the systematic design of the control systems. Modeling can be roughly divided into a method to construct a model from its essential principles (physical laws, etc.) and a method called system identification, in which a model is constructed from its target input / output data. In our laboratory, we are conducting research on the system identification method called subspace identification method. This method has been developed and spread since the 1990s. In the system identification method, it is required to provide not only the target model but also the necessary information for the control system design (e.g. robust control, trust area of ​​the model). We study the model accuracy analysis method and the parameter estimation method for that purpose.

Achievements are listed here