So as the observant of you will be aware, I am “trapped” in New
Zealand over Christmas this year, as I can’t leave and get back into the
Country. Whilst the situation is not ideal, it does present a unique
opportunity to get some projects done I’ve been putting off for quite
some time.
I have approximately a month (30 days) to get various projects done.
Here is a list of projects I am over-optimistically planning to get done
over this holiday:
Finish journal [7 days] – A journal I am currently working
on really needs to be finished, there is no way I can finish this
Christmas period before completing it - in fact this must be complete by
Christmas day.
PhD experiments [7 days] – A considerable amount of
computation becomes available over Christmas and I really want to get
some experiments running over this time. Specifically, I want to get
some multi-agent search experiments up and running.
Underwater ROV [7 days] – I need to implement a wireless
communication protocol for an underwater robot. This needs to be
completed soon.
IoT experiments [7 days] – I need to setup, design,
implement and demonstrate a new security protocol for IoT for an up and
coming paper. This needs to be completed ASAP.
Christmas [3 days] – Christmas day is most definitely
spoken for and I suspect heavily that Boxing day is also. I’m not
entirely sure about New Years Eve, but I do plan to celebrate 2021
coming in. I have made plans to meet several people over this time, so
there is quite some commitment.
Explore New Zealand [3 days] – I plan to spend about
$200-$300 and see some of New Zealand after - I got an awesome invite by
some friends to travel with them.
AI paper [3 days] – This is a project I’ve started with a
RoboCup friend based on the future of AI in RoboCup, specifically
learning. I suspect this will be about 3 more days of work spread out
over the Christmas period.
Website audio [1 day] – This project is to create a
podcast-esque auto-generated reading of the article pages.
NOTE: This is already completed.
Pinetime heartrate [3 days] – This is to take
infinitime OS and add a heart-rate sensing functionality, so
that it can be used as a proper smart watch. This involves setting up
the development environment so that I can compile code, writing an I2C
driver for the heart rate sensor and getting the user interface to spit
out sane data. NOTE: I have already worked on this for
a day.
Oakwm improvements [3 days] – There are a series of
improvements that are required to make the OS more usable daily, such as
power state (multiple sources), configuration options and just setting
up some basic software to work correctly.
Vision experiments [7 days] – I am working on YOLOv2, have
some ideas for experiments I want to run and need to setup a system to
run them. This will likely take many weeks to complete, but the process
should be started. I unfortunately missed a publication deadline on this
already simply because I ran out of time to implement my ideas.
Motion experiments [7 days] – I designed an experiment a
while back for self-motivated motion learning (towards a motion engine)
and really need to finish coding up the experiment environment and
actually code the behaviour. Maybe there is some fundamental problem,
but better to find out now rather than later.
PCB design [3 days] – Our robot really needs a power PCB to
be designed and build. These designs can then be sent off to
manufacturing in China and we can begin the process of iterating on the
design.
Dead social [7 days] – This is a project I started a few
weeks ago to make a very basic social media system for this website,
allowing we to write short content remotely. The idea is that I could
have a brain fart and log into the site from my mobile to write about
it. NOTE: I have already spent 2 days on this.
Nvidia board [3 days] – This is our new computing board and
it really needs to have our source code compiled for it in the
development environment. This will be quite annoying to setup
automatically as Nvidia put their own restrictions on how their
libraries can be used.
As you can already see, the number of days I want to get completed in
the month already exceed the amount of time I have, so let’s see… I want
to get 57 days worth of stuff done in 30 days!
Stretch and reach:
IRC notifications – I want to be able to have a bot post
RSS feed updates into an IRC room. This could be feeds from source
repositories, news websites, etc. This would really add value to some of
the rooms I’m currently part of.
Motor design – I have for a while now wanted to explore
motor design for the robots. Something that stopped me doing so was
getting the motors to conform to a specific dimension, but now I am
tempted to say “hell to that” and just do it to which ever size, just to
see a working prototype. In any case we eventually move towards larger
robots.
Programming book – A while back I started a programming
book about building ultra small OSes - I really want to see this
complete (and may even generate revenue), but I simply haven’t had the
time so far.
Gameboy game – One plan is to build a Nintendo game boy
game in C, based on a plot I wrote a few months back for an anime. I
really think this idea will be quite nice, but I could easily sink an
entire year into it alone.
Much less important:
Watch TV series – I really want to watch some TV series I
have put off all year. I somehow have managed to fit in watching the
latest seasons of Star Trek Discovery and The Mandalorian recently, but
this is pretty much it.
As you can see, as per normal I have been ridiculously
over-optimistic in my time planning. I will push myself and we’ll
hopefully revisit this list to get an idea of exactly how bad my time
planning really is.
Believe it or not, this is a massively reduced version of things I am
currently thinking about.
Coffee Space 
