Re: RULES or SPECIFICATIONS?

["Jerry Niman" <J.Niman@mmu.ac.uk>]

Tony,

I think your analogies are incorrect.

It matters a lot to a micromouse how the maze is built, it matters very little to a football or
pool player what the exact dimensions of the pitch or table are.

I agree that it is unrealistic to expect all mazes to be identical in dimension, but the rules
should specify the tolerance and the important characteristics of the construction as far as a mouse
is concerned.

As I recall, and I could be wrong here, the rules of football do lay down pitch dimensions...

Jerry


Jerry Niman                                    Tel    +44 (0)161-247 1474
Head of Information Systems                    Email  J.Niman@mmu.ac.uk
the Manchester Metropolitan University         Mobile +44 (0)7770 638104


>>> tony.wilcox@tic.ac.uk 30 Jun 04 22:02:01 >>>
Duncan and all ...
 
There is a big difference between rules of competition, and specifications for the maze, timing
gear etc. This is one area where I think  IEE/RHUL/TIC and APEC have got it wrong.
 
Analogies:  
You can play exactly the same game of pool on any number of different sized tables, with different
sized balls and pockets.
You can play exactly the same game of football on any number of pitches with different surfaces and
lighting conditions.
 
We can formulate a very well-bounded set of COMPETITION RULES, which are clear and unambiguous.
There isn't really a lot of work to do to the current set(s).
 
But as for the physical elements - the maze, timing gear, environment ...
 
We have a known FORMAT for the maze - 16*16 with 180mm*180mm cells, using 12mm*50mm walls, and
12mm*12mm pegs.
 
Other than that ... 
 
We have a guide as to colours - red,  white  and matt black. 
We have tolerances specified for dimensions. 
We have wood specified for the base - but no mention of material for the walls and pegs. 
We have suggestions about  the location of optical timing gear. 
 
By their very nature, mazes are going to vary - if you want to specify the dimensions and
construction of the maze such that every one is identical - you would need a production line, a lot
of cash, and a host of international standard specifications to boot.
 
 It is the task of the mouse designer to ensure that his mouse can cope with small variations in
reflectivity, colour, ambient lighting etc.in the same way that it is the designer's task to cope
with peg-slots and wall gaps. 
 
My abstract view of micromouse :- autonomous vehicles that use their senses to best advantage to
solve a problem presented in a defined, but imperfect, environment.
 
We can only do our best to provide as near-perfect a maze as our resources will allow. The mouse
designer does his best to cope with the allowed variations.
 
So let's separate the RULES from the SPECIFICATIONS ... if we work on smaller bits, we may get to a
solution faster!
 
Regards
 
Tony

	-----Original Message----- 
	From: owner-micromouse@cs.rhul.ac.uk on behalf of Duncan Louttit 
	Sent: Wed 6/30/2004 9:54 AM 
	To: micromouse@cs.rhul.ac.uk 
	Cc: 
	Subject: RE: Rules
	
	

	Hi Everyone,
	
	        The problem is that the current rules don't reflect reality. For example,
	I don't see how you can use posts with 3mm wide slots 3mm deep and still be
	within the spec. of 1mm maximum gap in the walls and a 5% tolerancs on
	dimensions.
	
	        What has happened is that the competition has changed but the rules
	haven't. If you design a mouse according to the rules and it fails because
	the maze doesn't comply with them, you will be put off. What is required is
	a clause by clause evaluation to ensure that the rules represent the
	competition as it is run so that we don't confuse new competitors.
	
	        It is vital that the rules be updated by someone (or team) who have recent
	experience in building mice. We need to ensure that the information
	necessary to design a mouse is provided by the rules. Implying that the
	size of handicaps is insignificant is not good enough. If someone is
	spending hundreds or thousands of hours working on a mouse, they at least
	need to know how it will be judged. Modern competitive mice are highly
	optimised and tiny details of the maze or the rules may become significant.
	
	        On the same lines, it is also vital that rule changes are made in a timely
	manner. I would suggest that any rule changes for next year's competition
	should be finalised not later than the end of September so that builders
	have plenty of time to deal with the revised rules.
	
	        Regards,
	
	        Duncan