How the lab works

When you choose a lab, the research topic matters — but so does the way it works day to day. This page describes how GDAC Lab is run.

What the lab values

The aim is to work control theory out properly and put it on the record, as conference talks and papers. We do not stop at “it ran” or “the numbers came out”: what matters is making clear what is new, and why the method works.

At the same time, laying out an environment in which ordinary effort produces a real result is the supervisor’s responsibility too. Excessive workloads and unreasonable demands are not the premise here. Time goes into the choice of topic, the reading and the daily discussion for exactly that reason.

Experiments on real hardware matter as well. Tools have made it quicker than it used to be to work a theory through or run a simulation, which raises the value of what you find out by building something and switching it on. Running the hardware is not the point in itself, though. What counts is experimenting with a proper grasp of the theory, so that when it misbehaves you can pin down for yourself whether the theory or the hardware is at fault.

In research, the answer is almost never known at the outset. You look things up, cut a problem into pieces small enough to attack, try them, and explain what came out. That way of working outlasts any particular topic.

Research is not easy, and it does not always go as planned. Even so, working at a hard problem and becoming able to do something you could not do before is the interesting part. One goal is that nobody leaves with regrets, and that they leave thinking the research they did here was worth doing.

Life in the lab

Students come to the lab on weekdays as a rule, because being able to see the supervisor and the other members, and talk things over, matters too. There is no core time, though. If you need a day off, that is fine as long as you say so in advance.

Research reports and seminars are held regularly. A report is not only for sharing progress: it is where we sort out what is giving trouble and what comes next. Students who have just started get close support from the supervisor and from senior students.

You are not expected to sit on a problem alone. Thinking it through yourself matters, but when you are stuck, just ask.

For job hunting, tell me once your plans firm up. If I know when you will be away and roughly for how long, the research schedule can be arranged around it. Rather than settling it in advance, we talk it over as it comes up and find an arrangement that lets both move.

How research topics are chosen

Topics start from a broad direction suggested by the supervisor and take shape through discussion.

A fourth-year project begins with understanding control theory the lab has already built and checking it in simulation and on real hardware. Building a new control theory single-handedly is not asked for. What counts instead is:

• understanding the theory you are working with well enough to explain it in your own words
• picking up the related background
• getting an experiment working yourself

As you get used to research, you take on a growing share of finding the weaknesses in existing methods and deciding what to try next. The aim, in the end, is that you can carry the research on your own.

Conferences and papers

Results are presented at domestic and international conferences and published as papers.

You do not need to know how to read or write a paper, or give a talk, when you start. Finding the literature, writing results up and presenting them are all taught here.

Past work is listed on the Publications page.

Joining the lab from abroad

If you are thinking of joining from outside Japan, email me at nakano@nitech.ac.jp with your CV, your transcript, and a short account of what you would like to work on. I will normally suggest an online meeting first, so we can both judge honestly whether the fit is right, and whether the topic you have in mind is one this lab can support well.

Admission goes through the university rather than the lab: the application routes and the entrance examination are set by Nagoya Institute of Technology, and the outcome is not mine to decide. What I can do is talk it through with you beforehand and point you at the right route.

Visits and contact

If you have questions about the research or about how the lab is run, get in touch at any time — write to me at nakano@nitech.ac.jp. You are welcome to visit; the location is on the Access page.

When enquiries arrive in a cluster, around the graduate entrance examination for instance, I may propose a few set times rather than answering each one separately.