High-Speed GA Pathfinding
Has anyone created a genetic algorithm which develops best paths toward an object. For example, say the AI is playing Age of Kings. It has maybe, 5 "hyperunits" which are invisible units that go faster than the speed of normal units, and are there to find good paths for the AI to go to. These hyperunits go at random directions, but eventually after a few generations, they narrow down a few good paths for actual AI units to take depending on certain criteria (fastest path, shortest path, least dangerous, most dangerous, etc.)
[email=ubc_wiskatos@hotmail.com" target="_blank" style="width: 10px; height: 10px; background: #fe7a21; overflow: hidden; display: block; margin-bottom: 2px;][/email]Wiskatosxp
GAs aren''t much use for creating paths in real-time for computer games. Too slow.
They do have their uses though. This is a link to a GA that finds paths in non-discrete environments.
http://www.ai-junkie.com/files/GA_Pathfinder.zip
My Website: ai-junkie.com | My Book: AI Techniques for Game Programming
They do have their uses though. This is a link to a GA that finds paths in non-discrete environments.
http://www.ai-junkie.com/files/GA_Pathfinder.zip
My Website: ai-junkie.com | My Book: AI Techniques for Game Programming
My Website: ai-junkie.com | My Books: 'Programming Game AI by Example' & 'AI Techniques for Game Programming'
I''ve often thought about playing with a GA to evolve an optimal heuristic for a given map. It might just be balancing a ratio between Manhattan and Pythagorean distance, for example. I wonder if that would ever be worth the effort.
I''d say it was worth experimenting with. It''d be interesting to see the results. Good idea.
My Website: ai-junkie.com | My Book: AI Techniques for Game Programming
My Website: ai-junkie.com | My Book: AI Techniques for Game Programming
My Website: ai-junkie.com | My Books: 'Programming Game AI by Example' & 'AI Techniques for Game Programming'
some guys over at the university if california (i think :/) mimplemented GA pathfinding in an unmaned underwater vehicle. There is an article at citeseer about it. They claim it works good, and the good thing about it is that it can easily recalculate a new path if an obstacle suddenly appears....
--Spencer
"Relax, this dragon is sleeping..."
--Spencer
"Relax, this dragon is sleeping..."
--Spencer"All in accordance with the prophecy..."
quote:
Original post by Spencer
some guys over at the university if california (i think :/) mimplemented GA pathfinding in an unmaned underwater vehicle.
Is that the same UAV that Gary Ogasawara was working on while he was at Berkeley? I believe he went on to Lockheed AI where he furthered his work on underwater vehicles.
Cheers,
Timkin
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
Recommended Tutorials
Advertisement