Skip to main content

Timeline for Obstacle bypass in 2d environment

Current License: CC BY-SA 3.0

9 events
when toggle format what by license comment
Jul 8, 2015 at 18:33 history edited Zachary Dow CC BY-SA 3.0
Dijkstra would be fairly inefficient given the requirements laid out in edits and comments.
Jul 8, 2015 at 17:48 comment added Zachary Dow Well, anything you throw at it that would be different than this map could potentially trip up whatever you make short of literally testing all paths. It may be overkill, but this reminds me of neural networks. You can essentially have your robot get smarter over time. Reminds me of MarI/O. youtube.com/watch?v=qv6UVOQ0F44 The mario kart version seems much like what you're doing: youtube.com/watch?v=S9Y_I9vY8Qw
Jul 8, 2015 at 17:47 comment added null @bux that would make for a hilarious scene in the Termintor 25 movie, with aged characters.
Jul 8, 2015 at 17:42 comment added bux Imagine object is a robot in your house. It can test every position like Dijkstra's algorithm but it will take two week to bypass your chair ^^ I add this precision to my question.
Jul 8, 2015 at 17:40 comment added Zachary Dow That doesn't matter. Look at an example of Dijkstra's algorithm in action. As long as it has the ability to analyze every move it can take, it shouldn't matter. upload.wikimedia.org/wikipedia/commons/2/23/…
Jul 8, 2015 at 17:38 comment added bux Object can't "know what it can or can't cross over" on all positions. It can know this only for near squares. It add this precision in my question.
Jul 8, 2015 at 17:36 comment added Zachary Dow @bux So it can know where it's objective is (It knows direction and distance, so it can know the location). It can know what it can or can't cross over. Why can't it use one of those algorithms then?
Jul 8, 2015 at 17:32 comment added bux Hi, according to my question we can't use an algorithm like A* or Dijkstra. we can only use information around object in movement.
Jul 8, 2015 at 17:24 history answered Zachary Dow CC BY-SA 3.0