Duncan Louttit wrote:
I have been giving some thought to the Rat race course. At the moment it is very difficult to make a course for testing purposes. When Creag and Steve made I.T. we ended up with lots of bits of hardboard hinged together. The only place we could lay it out was from the hall across two rooms!
It would appear to make sense to have a course that could be made from just one standard sheet of 2.4 metres by 1.2 metres material. If the total length including staging and braking areas is 7.2 metres or less and the width of one lane is 40 centimetres, a test track can be made from a single sheet cut into 3. The three pieces can be hinged together for storage. Alternatively, the track could be 4.8 metres long and 60 centimetres wide.
The reduction in length would also make it easier to find space to lay out the course.
A narrower track would encourage smaller mice and closer line tracking.
The staging area needs only to be 36 centimetres long to allow for the longest mouse allowed under the rules.
What proportion of the remaining length is for track and what is for braking would need to be determined.
Jerry Niman wrote:
Sounds like a good idea to me. I would think that the 7.2m x 0.4m option is closest to the spirit of the original idea.
The 2001 track has a finishing area 1.6m in length. This would leave a running length of 5.24m, as opposed to 7.3m for the 2001 track, not too drastic a reduction and worth it for the ease of constructing a test track.
The layout at the start should allow for an uninterrupted track centre line, as opposed to the 2001 track which had an uninterrupted start line. This meant that mice had to start with their sensors 'over' the start line to make sure they picked up the track centre line.
Chris Walker wrote:
Admittedly, laying out a test track for the Rat Race has always been a bit of a burden involving several sheets of MDF and furniture shifting. However, schools generally would probably find the room more easily than would competitors at home (classroom, corridor, hall, gym, playground, Head's office.......). Nonetheless, it does make sense to design the course around easily sourced materials. A few of my thoughts........
I'm a bit concerned about reducing the dimensions of the current course - I maintain that the Rat Race should be a challenge of power and control:
1) Shortening the track means the whole race is over and done with much quicker, possibly reducing spectator interest?
2) Less running distance implies lower average speeds - possibly reducing the excitement/daring factor? (Although few competitors are yet entering high speed Rats).
3) A shorter race time increases the significance of a good start - human reactions play a part here - is this significant/fair? cf. Techno Games rocket car racing.....
4) A shorter race time also increases the demand on accurate timing systems.
5) Reducing the track width to 0.4m implies a greater need for accurate line tracking (although most of last year's Rats tracked very accurately) - are we making things harder for schools' to achieve?
6) Talking numbers:
Current track (excluding staging): 7.3m (82%) running, 1.6m (18%) braking.
If we use Duncan's idea for a 7.2m course (minus 0.4m for staging), the same proportions would give 5.6m running, 1.2m braking.
This is a reduction of nearly 25%.
There's two ways of looking at this:
If we assume that the Rats accelerate constantly over the running length, then a 25% length reduction will only reduce their top speed by 13% (since v2 = 2as). The running time will also decrease by this factor. 13% doesn't seem such a big reduction.
However, since most of the Rats probably reach top speed within about 2m of the start, then the reduction factor is more significant.
Is this all getting a bit silly??! The bottom line is that we won't really know if it works until we try it..... I'm prepared to give it a go.
Duncan Louttit wrote:
If the left hand wall is made 50mm high, 12mm thick and painted white with a red top, then (slightly modified) wall-followers can enter. All they need to add is a kill switch operated by the end gantry.
More thought is needed about the start line/centre line interface. Should there be a gap between the start line and the centre line so that the mouse sensors never see the start line? How wide does the gap need to be? If the start line doesn't go all the way across, how can the judges check that the mouse is behind the line?
One solution may be to use a narrow start line. If it is only 0.5mm wide (drawn with a pen of some kind?), it would not upset many mice.
Jerry Niman wrote
If the start line is interrupted, we can easily check that the mouse is not over the start line by using the timing gear.
Yes, by borrowing my daughter's GCSE Systems and Control Technology revision notes, I have built a rat race timer controlled by light beams at each end.
The software will give a clear indication of whether the start line is clear or not (until the race starts, then it gives a 5 second penalty !).
As to how wide the gap should be, even an edge follower should be 'happy' with a gap equivalent to the center line width, 50mm.
The idea of a thin start line is also a good one.
Duncan Louttit wrote:
The gap in the start line would need to be a lot bigger than the width of the line. I.T. had an array of sensors that went all across the front. It could recover from any position that put one or more sensors on the line. On this basis the gap needs to be 35 centimetres to cope with the maximum dimension of the mouse. Not much line left on a track of 40cm width!
However, if the timing gear has a "good" light available for the competitor to see, then all that is needed is a guide line to help them get somewhere near. 2.5 cm each side may well be enough for this.
Will Sheaff wrote:
> 5) Reducing the track width to 0.4m implies a greater need for
> accurate line tracking (although most of last year's Rats tracked very
> accurately) - are we making things harder for schools' to achieve?
Well, not really. Most people must have realised that verminator was horribly overpowered, (and the steering tended to get wobbly towards the 7th meter of the track). Surely a more succesful (and reliable) rat would be one that had much more emphasis on steering, and a motor that reached the absolute top speed the steering could cope with.(so as to keep the weight down). Even at the current width, steering is still basically the most essential component, (and surely that was the idea of micromouses origionally: to challenge schools to build the precision robots?)
If it's decided to shrink the course, for easy of storage, etc, then at least leaving a 50mm line wouldn't require changes to any "suped up" ("souped up"?) formula 2 and three mice.
And finally, most of the rats only follow one edge of the line anyway, with a two-on, two-off sensor arrangement. The problem could come then, where the line becomes too narrow to fit the two sensors on it: very confusing for the robot.
Duncan Louttit wrote:
If the finish "gantry" is just two vertical poles, one on each side of the track, it would be impractical for a mouse to use them for braking. The mouse cannot stretch across both poles without being oversize and if it uses just one pole for physical braking, it will spin off the track.
This would require some minor redesign of Verminator to use a sideways mounted finish switch. Would this cause a problem?
Chris Walker wrote:
The Finish gantry has a certain aesthetic appeal - complete with chequered flag and micromouse logos - I'd be sorry to see it go.
Also, the height of the gantry is constant across the width of the track, so any braking-activation switches are bound to hit the gantry irrespective of where the Rat is across the width.
If a Rat relies on a vertical pole at the side of the track, then a braking switch would have to extend far enough sideways out of the Rat to ensure it hits the pole if the Rat should find itself at the opposite side of the track as it crosses the finish line. I can visualise a scenario where long braking switch arms might interfere with the Rat in the neighbouring lane. I think that this method is rather inelegant - I vote for keeping the gantry.
Suggest an addition to the rules along the lines of, "the gantry will withstand light touches (e.g. a microswitch lever) but should not be relied upon to slow down the Rat".