RouteMaster May 2013

3 things learned...

9thth May 2013

  1. Due to the weight of the mouse, there is a tendency for the momentum created in the fast spin to allow the mouse to continue spinning beyond it's stop point if another move is not performed straight away. If the mouse has to have a think or "feel" after the spin, make sure power is maintained to the steppers to effectively put on the hand brake until ready for the next move - or for a short period of time.
  2. Using the wheel chair design, I assumed that turning one wheel clockwise and the other anti-clockwise, the mouse would spin around the centre of the mouse. Well, apparently mine doesn't. In turning the 90 degrees, the angle is correct, but the mouse has drifted off the centre-line by a couple of mm. This means that as it moves away, it is no longer moving down the centre of the cell. I can only assume that the wheels are not equidistant from the centre although they appear to be more or less at the end of the motor axles.
  3. The last thing learned - don't undo the grub screw holding the wheels to the axle too far. I undid the grub so much that it came out, but without the correct took, I now can't get the angle on the grub to be able to put it back in without cross-threading! "One wheel on my wagon..." - Arggghhhh!


8thth May 2013

I tried to make some tyre covers from balloon elastic last night in a bit to try and get a bit more grip from the tyres. I wonder if the tyres are not gripping well and that is what is causing the mouse to perform poorly. Having struggled to learn the technique of applying the tyres, I fond that the new rubber was rubbing against something on the body or wiring. This caused the wheel not to rotate smoothly, so I have removed then for now.

Deciding that the mouse would probably never perform perfect movements, I then embarked on some fault-fixing routines. For example as the mouse has front and rear sensors, I can actually check to see if the mouse is sitting parallel to a wall - enabling a small spin routine to tidy up the position. That seems to work OK - all I need to do now is to remind myself how I ready the map from any position and direction and decide if there is a wall available to ping against! Duh!

The left wall steering routine appears to be working as the mouse is seen to be swerving left and right along a straight which shows the correcting routine is working. Just wish all the other moves were more accurate!


7thth May 2013 23:50

Phew! The power problem ended up being a short in the LCD module wiring. Wiggling that made everything come to life, so that unit has been removed for now. Feedback is still available from the RS232 cable link.

Here is a little video of the mouse (with a little assistance) solving the 5x5 maze.


Good News/Bad News

7thth May 2013 22:27

Good news is that the flood and direction code appear to work OK as RM2013 managed (with a little assistance) to solve and reach the end of my 5x5 test maze! Yipee!

Bad news - catastrophic power failure to the main processor board. The power comes from a 9v PP9 battery onto the board via a 2 pin connector. I'm hoping that it is a simple connection failure. If not, then I suspect it is the onboard power regulator - gulp! Let's hope it's the lead!!!


Blow the dust off!

6thth May 2013
Well, the dust has well and truly been blown off and preparations are being make for this season � starting with the Otley College �Big Day Out� workshop. The Workshop has an 11x11 maze, so we can fully test the mouse as what would be the �Centre� is covered by the maze. A full new bottom-up re-write for RM2013. (Well, re-using modules of pre-written code along with brand new code). The bank holiday has allowed some MicroMouse work as well as gardening etc etc! More to follow�