Greetings to all.

I'm looking for an advice or directions for my pathfinding problem.
Game has a continuos space with static and dynamic obstacles.

Link for image - http://www.gamedev.ru/files/images/pathfinding.jpg

Red color is used for static obstacles, and green for dynamic ones.
Game agent path should should be calculated without collisions with colored rectangles that are representing obstacles.

There is no need to find shortest path to target.
Amount of dynamic obstacles, i.e. agents, would be equal or greater (30+) than amount of static ones.

Looking for an advice, hint for an feasible algoritm or some direction, there i should look to solve this problem.
Thanks.

If the obstacles are as sparse as the image you posted, look into steering behaviors. That's probably all you need.

I'd agree with steering behaviors as far as avoiding dynamic obstacles, but since I assume that the dynamic obstacles are what you are moving through the map, I'd use Greedy Best First Search for the static obstacles.

Thanks for answers.
I learned about presented paths to complete the task, and chosen to go with potential field method (got pointed to it at another forum).


It solved all pressing issues with movement and presented nice side bonuses as a reward. If someone interested he can learn it here - http://aigamedev.com/open/tutorials/potential-fields/

Peace to all.