Friday 29 January 2016

w/c 25th Jan: Work, Interrupted

It's been quite a "bitty" week, this week. Lots of odds and ends. One-off lectures, installing the new version of FATKAT, reviewing a paper, giving feedback on a thesis, PhD supervision meetings, editting one of my own papers, signing off paperwork for a commercialisation bid, submitting exam and coursework marks... you get the picture.

But if I've been reminded of the importance of NOT having an open door policy. Open door policies sound great - just drop in, whenever! In practice they are non-committal. I was out of the office for various stretches each day, so if you were dropping by in the hope I'd be in, then you were probably out of luck. And then there's the problem that even when I am in the office, I'm doing things. One or two people dropping by is doable: but when people say "Do you have five minutes?" it's never just five minutes. And it breaks your flow. That "five minutes" probably costs you at least fifteen, in practice. And if you do this for three or four students, that's an hour of lost prep or marking, which means that someone else loses out down the line in the form of delayed feedback, or hurried marking. Few students would thank you for skimping on their work to spend time on another students. And if you do it for three or four students, where do you draw the line? Can you do it for thirty or a hundred students? Of course not - so where do you draw the line? If you'll do it for the first student who drops by, how do you say no to the twentieth?

The best solution I've found is office hours: to have fixed times when you're available, and insist students see you then. That way, you know when you'll be Interrupted, and students know when you're around.

The exception that I make is for pastoral care. Sometimes you do get a tutee who turns up to discuss the fact that there's a family crisis and they're ill and... you get the picture. For that, there's always time.

Anyway, on a lighter note, I had the pleasure of examining a thesis on Monday with David Franklin from the Computer-Based Learning lab at Cambridge (though he's soon to be of Munich, as of early next month!). Here at Leeds, we're big fans of the CBL group, so it was great to have a chance to get his perspective on the work we do. A good viva is a double pleasure: an interesting thesis to discuss, and getting to meet researchers with whom you might otherwise not cross paths.

Anyway: more of the same next week (except the viva), ticking the small items from my To Do list!

Monday 25 January 2016

w/c 18th Jan: Exams and Strategy


If last week was a week of making, this week has been a week of exam marking. My Mechanical Systems exam struck on Tuesday, and I’ve been marking the papers since. This is time consuming, but over the years I’ve learned how to set questions that are difficult to answer, but easy to mark. So, I’ve spent a lot of time with red pen, totting up marks, and keeping tabs on how the cohort perform on each question.

But you don’t want to know about my exam. What else has been going on this week? Well, I finished building the two new FATKAT manipluandums, and got started on revising the software. I’m making it more flexible, so that you can specify what each input is, rather than being bound by the default. This has meant writing a bunch of code to sort 2D arrays in LabVIEW, but I’m largely there.

Also, we had a Mechatronics and Robotics Strategy day, organised by the newyly-minted Professor Abbas Deghani (by which I mean, newly made a Professor), to discuss our priorities and strategies over the coming year. In particular, it was an opportunity for the new University Academic Fellows to show off their work. Very exciting – and very multidisciplinary. We had lawyers, economists and artists as well as engineers of all stripes. There’s lots to be excited about. For me, in particular, Mehmet Dogar’s work on robotic manipulation ties in very nicely with some of the modelling I aim to do on prehension.

2016 is already shaping up to be an exciting year!

Saturday 16 January 2016

w/c 11th Jan: a week in pieces



This is what I've been looking at for much of the week: nuts, bolts and Lego! That's because FATKAT manipulandum 2.0 is coming together. The design is done, ready for final build next week. Final for this iteration, anyway. It's a Frankenstein's Monster made of off-the-shelf components. I'm confident now that we can get a slimmer, more elegant, laser cut shell on the same chassis, but that's for the future. Right now, the emphasis is having something good enough ready for Will to start testing on the 25th. The watchword (er, watchphrase? Watchslogan?) of the engineer is "fit for purpose".

While that's been a big chunk of my week, it hasn't been all of it. I've been knocking up a foot force measuring system using an Arduino for Callum Mole. I've been marking interim reports, and I've been reading theses and organising vivas. I got through two theses this week (albeit corrected ones that had already been examined), and have arranged not one but THREE upcoming vivas that I will be examining in the next two months. Theses are big documents, but the nature of doctoral research at least means they are always interesting! On to next week, when I've an exam to mark, and sadly I won't be at the LUDI meeting in Aosta...

Saturday 9 January 2016

w/c 4th January: It's back to work we go...



Well, this has been a busy, busy week. By its nature, returning to work after a holiday means lots of new things get started (all the stuff there was no point starting before), and that means that while lots of things go onto the To Do list, very few get to a point where they can be crossed off*. Such has been the case this week. Apart from anything else, the first morning was taken up with drawing up my To Do list and planning out my diary. Always a frustrating exercise, when you feel like you should be doing things, but I long ago learned the hard way that if you don't invest the time to get organised you'll be in chaos for the rest of the term. So, time well spent, and it has enabled me to draw up a priority list and focus my time. And there's been plenty to focus on.

A big grant proposal went in to EPSRC this week - I wasn't leading, but it meant planning in time to review and comment on the draft in enough time for it to be returned.

The new FATKAT manipulandum has reached a new prototype stage, which has meant playing around with bolts, nuts, lockwashers, drain connectors(!), and Lego (naturally). That's it at the top of the page. It's very much a prototype, still, and it looks rough, but I think this is what I'll stick with for the first batch of testing. It isn't pretty, but it works - and investing a load of time in laser cutting or 3D printing parts just runs the risk of delaying Will Shaw's data collection. Plus Lego gives us the flexibility to attach a Delsys accelerometer to it, without the need for making a permanent difference to the design.

The WHISPER prototype has continued to develop - after an initial demonstration and discussion with Bryan Matthews from ITS, we've got some good ideas for initial feedback signals, and I'll be off to implement them next week.

Finally, we've gathered some aiming data from CKAT, to test some of my theories about grip control and aiming movements. That's entailed writing some new programmes to translate CKAT data into the form that I use for processing Optotrak data. I've managed that, and got it all processed - the results give me cause to be optimistic, but the current sample size (N=1!) isn't nearly enough to rule out chance, so now that we've got that it's off to gather more datasets to test and interpret.

And then there are reports to mark, PhD students to supervise, a bid for commercialisation funds to redraft, and three theses that need reading before I examine them in the spring (and one corrected thesis that needs reading as well!). So, lots of new things on the To Do list, not many crossed off - but lots of progress despite that. It's nice to feel like you're building up a bit of momentum...

Friday 8 January 2016

2016: Year in Preview


I have mentioned before that I don't make New Year's Resolutions, but the start of a new year is a good point to take stock, before exams and term hit. So in this post, I wanted to take a look at what's on my horizon in 2016, and what I hope to have achieved by the year end. Let's break that down into three strands: research, making and the blog.

Research: There are a few strands going on here. First and foremost, there's an obvious one: deliver WHISPER successfully. That stands to reason: we've taken the King's Shilling (well, the DfT's permission to bill them for the cost of the project), and now we have to make good on our promises. It's not trivial, it won't be easy, but it will be interesting.

Then there are research proposals. I'm aiming to get five submitted this year. Last year, I had set myself the challenge of working on more collaborative grants, and it really paid off. This year, I've got five grants earmarked for submission (including one submitted in the time it took me to write this post - so I'm on my way!).

Then there are research publications. Let's go with two high quality journal papers (the 3*/4* REF type, not just the basic science). I've got one on in draft - and an idea for another, if the data goes our way. That's a big if, of course: but such is the way of science. Refuting your hunches is important, but you only get the real attention when they pay off. But that's out of my hands: let's just focus on getting this published.

Finally, there's FATKAT. This is going to be a big year for FATKAT, and that means I'm currently foot to the floor making the new hardware and polishing up the software ready for data gathering at the end of the month. Real data gathering - not just pilot data or validation studies. If 2014 was the year FATKAT was born, 2015 was the year of putting it through its paces: and 2016 will be the year of actually doing research with it.

Making: 2015 was a good year on the making front, I think. I didn't really make anything outside work (unless you count assembling kits!), but I got myself signed up with Leeds Hackspace, I got my Raspberry Pi setup finalised, and PiCam working, and I played around with Onion Omegas, Arduinos, CodeBugs, Soldering  Kit a Month and Flotilla. All good fun, and a good way of sharpening up the old skills. This year, though I definitely want to get some projects actually worked through - so let's set three goals. Firstly, I'd like to get hands on with the laser cutter and the 3D printer at the Hackspace. So far, everything I've done has been assembly and programming, and this year I want to explore these processes more. Secondly, I want to get OpenCV running in conjunction with my PiCam - at the bare minimum I want to have the PiCam setup self contained and portable enough that I can do a time lapse video without it plugged into the mains. Finally, I'd like to have a go at making my own interactive Christmas decoration. I know it's not much, but I figure it gives me a whole year to actually design something fancy (not just a paper chain, or cardboard cutouts...).

Blog: Finally, I'd like to get the blog up and running rather better. Weekly updates are good (even if they don't end up being exactly weekly), because they show I'm still here and the blog is alive, even when I'm busy. But few people are that interested in my day-to-day activities, and I originally intended this to be a place for thinking out loud, which I do very little of here. So, this year, I'd like firstly to keep up the regular updates, but also try to get back to putting in some additional posts on issues around the Engineering Imagination, my making experiences, and the papers I read - a chance to draw my thoughts together. We'll see how that goes. Finally, I want to make better use of tags and images on this blog - walls of text aren't very inviting.

So to summarise, in 2016 I would like to:

1. Complete WHISPER;
2. Submit five research proposals;
3. Get research results from FATKAT;
4. Validate or refute my grip model;
5. Get two high-quality journal papers submitted;
6. Use the laser cutter and 3D printer down at Leeds Hackspace;
7. Get OpenCV and timelapse video working on my PiCAM;
8. Continue weekly(ish) updates on the blog;
9. Include some non-weekly update posts; and
10. Use tags and images more on the blog.

I mean, how hard can it be? Stay tuned!