1) Replacing the accelerometers - the prototype had two different models of accelerometer (thanks to the need for a last minute field replacement when the demoing to the International Spinal Research Trust), so I wanted to install two that we knew to be identical. It makes the processing software a lot easier to maintain, since the same pins refer to the same axes for both hands!
2) Replacing the protoboard with soldered circuitry - I'd used a protoboard for the amplifier circuitry (needed to get a decent resolution out of the force sensors), because I didn't know what gain we would need, and wanted to be able to easily change the resistors, without resorting to a variable resistor (that might then be inadvertently adjusted, etc). The protoboard was a real weak link, though - wires popped out, and needed replacing. Fine when I'm on hand to plug them back in, but not really sustainable if FATKAT is to be used by other people. Then I had the brainwave of using screw terminals to attach the resistors, so they could still be changed without the need to undo soldering, and so it was - a more permanent, soldered version of the circuitry was put together:
3) Trying the new calibration procedure and rewriting the calibration code in LabVIEW - this now seems to be working. My hope is that we'll only need to calibrate when hardware changes are made, but for now, the calibration process is a protection against wires being plugged in the wrong way round.
4) Having a few practice runs - we tried a few lifts each with 200g weights, and a few with 400g weights, then we each tried a slip force run (basically, picking the manipulandum up, holding it for a moment, then gradually releasing the force until it slips from the hand - this tells you the minimum force required to generate enough friction to hold it). The force trace is below:
It's coming out nice - you can see Will grips a lot harder than I do, but the slip force (last point before the force nose dives as the force sensor slips out of the hand) is very similar, which is what you'd expect.
And that's really been most of the week. The other big news was a PACLab strategy meeting, where we discussed global priorities, and how we wanted to take our various streams of work forwards and (just as important) how we communicate that work internally and externally. We haven't yet arrived at a finalised plan, but we're off to a good start. The three streams (Surgery, Developmental Research, and Rehabilitation/Mobility) are getting their websites planned out, and we'll be doing some technology roadmaps in the near future to look at what engineering Pete Culmer and I need to do to enable the next round of 4* REF outputs in time for REF2020.
I'm looking at grasp planning literature to underpin a paper that Mark Mon-Williams, Rachel Coats and I are writing.
I've also booked my flights for the next LUDI meeting - now confirmed for the first week of June in Limassol, Cyprus.
Last of all (outside work, admittedly - but relevant to the "Engineering Imagination"), I've been down at Leeds Hackspace, building my Raspberry Pi Camera system (complete with TFT screen. It's looking good. No preview shows as yet, so the aiming is a bit hit and miss, as you can see - but I'm well on my way to time-lapse photography, and some fresh computer vision projects...
Put like that, I actually got quite a lot done this week. Maybe this weekly review stuff will really work! Anyway, next week will be a lot less productive - Easter Bank Holiday, annual leave to look after my daughter (who's off school) will eat into most of the time. So I don't intend to start anything new - there are a bunch of things coming onto my desk (or back onto my desk) the week after, but for now, it's all about tidying up FATKAT, working on that paper, and brushing up the Pathways to Impact for my Fellowship application. Let's see how that goes...
No comments:
Post a Comment