System Control Theory and Networked Robotics
Uploaded on Jan 19, 2010
The Namerikawa Lab is researching networked robotics, primarily through system control theory. Networked robotics is a control system that enables multiple robots, connected via a computer network, to exchange information and move autonomously. This is achieved by creating decentralized algorithms mathematically that enable the robots to move in formation, coordinate, and be controlled in cooperation.
Q.Our research concerns system control theory and networked robotics. System control theory is the basis of this research; we study mathematically how to control dynamical systems reasonably and efficiently. The keyword here is feedback control. Feedback is utilized a lot in everyday life, and in our research, we study theoretically how we can create feedback architecture based on various physical models and mathematics.
Within the field of networked robotics, the Namerikawa Lab focuses on multi-agent systems. If a system consists of multiple agents, until now, operators have needed to give instructions to all the agents. But now, if instructions are given to just one agent in a multi-agent system, the agents can exchange information, and use it to arrange themselves in more advanced formations autonomously.
This research is expected to be used for autonomous following control of platoons of vehicles on highways and precision controlled spacecrafts and satellites flying in formation.
The Namerikawa Lab is also doing R&D on tele-operation systems. If feedback control can be utilized to remotely manipulate very large or small objects that cant be handled by people directly, this will make it safe to do remote surgery, or operate machinery in nuclear facilities. Recently, the team has been thinking of applying the system control knowledge and know-how its developed so far to controlling the Earths environment. The challenge is to control large-scale systems, in order to achieve green energy and reduce greenhouse gas emission.
Q.Control is an interdisciplinary field. Ive mentioned all sorts of topics, including nuclear energy and medicine, and we consider all these things as dynamical systems. So rather than specializing in physical phenomena, we study universal principles to control dynamical systems. Recently, weve been working on even larger-scale systems; specifically, weve a project on forecasting demand for electric power. And we also need to develop sensor agents to measure the Earths environment, which is an even larger system. The idea is that mobile robots will move autonomously over a wide area, and measure environmental information like CO2, temperature, and humidity to create a large database. The database might be used to predict how the Earths environment will change in the future, and what we need to do about it. So in our research, wed like to find ways of creating this kind of feedback.
The Namerikawa Lab investigates ways to control networked robotics, and even enormous systems like the Earths environment. This R&D will continue to advance, with the aim of ensuring an even brighter future through system control.