Jump to content

  • Log In with Google      Sign In   
  • Create Account

A* in Visual Basic?


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
3 replies to this topic

#1 Phyxx   Members   -  Reputation: 138

Like
Likes
Like

Posted 04 June 2000 - 07:13 PM

I am looking for a good path finding algorithm and a* seems to be my best bet. Unfortunately I really don''t have the time to sit down and convert the c and java code on the net into VB. Is there anyone out there who has implemented a* in visual basic (or who wants to implement it for me )?

Sponsor:

#2 StrategicAlliance   Members   -  Reputation: 122

Like
Likes
Like

Posted 04 June 2000 - 11:33 PM

Hey,

You might try out my VB-program on motion planning.
It''s open-source (well, the algorithm code is) and though I''ve probably not implemented a strict A*, there are several close variants in there. You could use one of these as a start-off point for your own one.
Let me know If you found them useful.

It''s in the experiment section on http://www.softline.be/StrategicAlliance/RUG/MotionPlanning/



******************************
Stefan Baert

On the day we create intelligence and consciousness, mankind becomes God.
On the day we create intelligence and consciousness, mankind becomes obsolete...
******************************

#3 Phyxx   Members   -  Reputation: 138

Like
Likes
Like

Posted 05 June 2000 - 03:14 PM

Yeah, it looks good... but how do I use it? You have all these sections:

Data
Module
Dijkstra (Visibility graph)
Roadmaps
Cell Decomposition
Potential 1 (Trial & Error)
Potential 2 (Best-first)
Potential 3 (Wavefront-expansion)
Potential 4 (Skeletal build-up)

What are they (modules, forms, class modules)? How do I actually put them into VB, and get a path?

I assumed that everything other than the potentials were the program itself, with the potnetials being demonstartions of this code.

It just looks a whole lot more complicated than the source code I have seen for other path finding algorithms. And the fact that everything is in dutch doesn''t help .

#4 StrategicAlliance   Members   -  Reputation: 122

Like
Likes
Like

Posted 06 June 2000 - 10:45 PM

I know the site may seem a little complicated, but I''ll explain a few things.

If you''ve seen the source at my site, then you''ve been browsing through my .BAS files from the VB project. It contains all the ''theoretical'' work. I didn''t include the code from the forms on the site.
The different sections all use a different technique to realize a pathfinding algorithm.

The potential-methods are actually 3 different methods making use of potential fields. I think Dijkstra is closest to A* and cell decomposition is a technique that splits the map into similar coherent regions and reduces the problem to finding a path between the regions instead of the whole map.

Roadmaps was an attempt to use random-waypoint generation to find a path in a quick way without exploring the whole map to find the quickest best path.

I would like to stress however that I wrote this program to support the report we had to make for the course. The code may be a little messy, because the primary object was to let colleagues in my year actually SEE what the methods do.

If you''re still interested and you send me a mail to my private account (see profile), I''ll send you the whole project (source code, forms,...), so you can run it from within VB...

Thanks for your interest!


******************************
Stefan Baert

On the day we create intelligence and consciousness, mankind becomes God.
On the day we create intelligence and consciousness, mankind becomes obsolete...
******************************




Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS