| July
23, 2004- Quicksilver
Quicksilver is my third real robot, and I’ve technically branded
it a “Multi-purpose autonomous robotic research platform.”
Basically, it’s a robot that I can use for lot’s of different
things.
This robot is part of my Year 10 IT second semester major project, and
I am designing it this term (term 2, semester 1) in Design. This means
I have just over half a year from start to finish. This is good, because
Munroe has taken 1 and 1/4 years and it still isn't finished! This robot
will also have a more structured timeline.
Progress
as of 11th May 2004:
Design is still in early stages, but has got off to a good start and a
definite design will be finished at the end of this school term and may
be published on this website.
23rd
June:
I have obtained an old Pentium 1 laptop for programming. It has a modest
64mb of ram and at 200 MHz it is a fairly modest machine, but it was a
bargain (free!) and it has the serial port that my newer Toshiba lacks.
Big shout out to my school for donating it.
2nd
July:
All parts are now ordered. I was hoping to get the IR proximity sensors
before holidays, but they’ll be here when school goes back.
20th
July:
IR has come as I expected from Acroname inc. robotics, and this week is
the start of our IT major project. It will be devoted mainly to planning
and organization, but next week will be the start of construction.
16th
October
I
have been very busy lately working on… well a lot of things. I have
too many projects going at once! Quicksilver has been at the front of
these projects. Significant programming has been done, and I am on track
with my timeline to have it finished in 5 weeks time. The core of the
robot’ operation
and it’s various indications is finished, and I have done a few
things to get to this stage:
- Added a floor sensor (same Sharp IR sensor as the two front ones).
- Fixed the reversing routine.
- Added a “boot up” sequence.
- Changed the rollover routine. It now has 12 second (approx) SLEEP periods,
with a 50ms LED flash in between. This is handy to know if it is still
on or not.
- Tweaked the front sensors in code and physically, to have less of a
“blind spot” at the front.
Forget
about wireless: I hate it. I am not normally good at getting things to
work without wires, and this time has been no exception. It was harder
than pulling teeth and even at the end still didn't work properly. If
I ever do get wireless comms to work, I will let everyone know! - It really
is less fun than drowning.
1st February 2005
It's been a while since I updated, but my major project period has finished
and I have presented the finished product to the rest of the class!
You may be able to pick the horizontal headers
on the interfacing board - this is wired up to accept a wireless modem
as detailed in Silicon Chip, but due to faulty PIC chips on the TX board,
it has never worked properly and I have left it off for the presentation.
When I have more money, I'd like to get some data transceivers from Oatley
so I can remote control it as well. The thermistor is wired to a voltage
divider on my only spare ADC (I chose analogue IRPs so I can tweak the
distance quickly) and I can get useful temp readings.
You may have noticed the IRP on the front - this
detects the floor, and via some clever programming it can't drive off
the edge of a table anymore! It only just reverses in time, and is still
the most popular feature on the bot.
The twin motor gearbox lost a few teeth on one
of the output cogs 4 days before presentation (leaving one side effectively
useless) so due to the lack of spares I had to buy another twin motor
gearbox and use the cog from there. I could still make one useful gearboox
with the leftovers, though.
Specs:
-Picaxe 28X microcontroller
-Picaxe 28X project board w/L293D dual motor controller
-Sharp GPD12 Analogue IR proximity sensors (x3)
-Mercury tilt sensor
-Temperature sensor
Chassis:
-Tamiya universal base plate
-Tamiya track and wheel set (this is a great package as the layout of
the tracks can still be changed)
-Tamiya twin motor gearbox
|