Talk:Micromouse

When you said "far east" (changed to East Asia now), do you mean Japan? Please be specific. --Menchi 12:48 13 Jul 2003 (UTC)


 * I mean Japan, South Korea and Singapore.


 * I found links to events in Ireland and Austrailia too, so changed it to say it runs worldwide. I changed the link from Wayne's site to A personal website as the name Wayne wouldn't mean anything to many viewing the page. Angela 19:25 13 Jul 2003 (UTC)

To User:BAxelrod:

I am curious. In the revision to the micromouse entry that you made on Oct 8th 2004, you state that the A* is one of the more popular search strategies.

It is my understanding that most mice have much simpler algorithms based on a straightforward flooding of the maze in the style of Lee-Moore.

I would be very interested in finding out about any mice that use A*

Pete Harrison

Topknot 19:23, 22 Nov 2004 (UTC)

To User:Topknot:

I think i was mistaken, because it looks like the Lee-Moore maze flooding is what i was refering to. This is not exactly my area of expertise, so if you know better, you can certainly change the page. However, i think that the Lee-Moore algorithm can be considered a simplified version of A*. (an examination of the external links on the A* page might make this clear). This simplified version has a very bad heuristic and a uniform path cost. It is not inconceivable that a mouse use A* especially when the locations of the start and goal states are known. This would allow a reasonable heuristic to be formed and would decrease search time. But in these competitions, search time is usually unlimited or very long so the overhead is probably not worth it. If you feel that the A* remark should be replaced with Lee-Moore, then please do so.

--BAxelrod 20:48, 24 Nov 2004 (UTC)

As far as I can tell, in Singapore, and I suspect this being likely in most other countries, micromouses typically use a simple rule-based algorithm. Micromouses are limited in computing capability and code size, the effort needed to implement a more involved algorithm is usually not a worthy tradeoff. The time gained is usually superficial as the standard for calculating runtime is largely based on the speed of the fastrun, and only a fraction of the search-time is considered. Besides, micromouse handlers often want to locate the best path, instead of just settling for "any path".

--Shasderias 12:36, 10 July 2005 (UTC)

Italic texthi this is ajay kumar —Preceding unsigned comment added by 210.212.208.226 (talk) 08:38, 6 February 2008 (UTC)

Needs update for the run times.

6 to 7 seconds!? BAH!

http://spectrum.ieee.org/automaton/robotics/diy/meet-the-new-worlds-fastest-micromouse

Brandon.irwin (talk) 01:21, 22 November 2011 (UTC)

Micromice?
Micromouses, shurely? David Lloyd-Jones (talk) 19:27, 20 September 2018 (UTC)


 * I'm pleased to find that Shasderias, above, agrees with me. In the IBM literature, btw, the things you push a cursor around with on screen are mouses.


 * David Lloyd-Jones (talk) 19:31, 20 September 2018 (UTC)