Skip to content

Latest commit

 

History

History
22 lines (10 loc) · 1.11 KB

README.md

File metadata and controls

22 lines (10 loc) · 1.11 KB

AntPursuitProblem, Swarm intelligence

Why the ant trails look so straight and nice? A thesis project. This is a simulation in Julia in which we can see that with a simple local interaction we obtain a global optimal solution for the problem of finding a shortest path. Can be used for example in robotics, using a single agent/robot that find a shortest path inside of a partially observable enviroment would be expensive and hard. Instead we can use (lots) easy,cheap agents to find the shortest path.

bruckstein_modello.mp4

This is my contribution in which the ants are able to deal with obstacles. To do that they use an heuristic which is the (white) ant that behaves like a "ghost", the other ants try to follow the "ghost" ant but they need to deal with the obstacle. Perserving the property of finding the shortest path

modello_ostacolo.1.mp4