search

DanielPerezJensen / mapc-uva

The UVA's entry into the Multi Agent Programming Contest
https://multiagentcontest.org/
1 stars 0 forks source link

There needs to be a way to tag agents with an ID #52

Open YassinAbdelrahman opened 4 years ago

YassinAbdelrahman commented 4 years ago

Tagging agents, allies and enemies, with an ID will improve tracking their movement.

gsileno commented 4 years ago

this is definitively a good idea.

On Sun, May 31, 2020 at 7:16 PM Yassin Abdelrahman notifications@github.com wrote:

Tagging agents, allies and enemies, with an ID will improve tracking their movement.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/DanielPerezJensen/mapc-uva/issues/52, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJFMVGCYH76C6JJMDLKL53RUKGGHANCNFSM4NPHY5BA .

DanielPerezJensen commented 4 years ago

This would be easy on a per-agent basis I think, simply storing an increment and assisgning it to each new agent we meet. But when merging, it could perhaps introduce some problems? @dorian4840

dorian4840 commented 4 years ago

This would be easy on a per-agent basis I think, simply storing an increment and assisgning it to each new agent we meet. But when merging, it could perhaps introduce some problems? @dorian4840

If each agent individually keeps track of the agents is sees, than you would simply have to change the coordinates on which you see them while merging. So that part wouldn't be difficult.

And once friendly agents see each other they will merge their graph so there is no need in tracking allies l.

DanielPerezJensen commented 4 years ago

The hard part will then be that we have to change update_graph() to facilitate this. Is it possible to assign a variable to an agent on a node and then have that variable move around with it, or would this have to be handled in update_graph()?

dorian4840 commented 4 years ago

That would be possible but I think it's easier to keep track of a list of agents and their location. Then every new step you create a new list of agent locations and assign them to the agent which most likely move to that new location.

DanielPerezJensen commented 4 years ago

Any work done on this? @YassinAbdelrahman

YassinAbdelrahman commented 4 years ago

It was not needed anymore for the attacker-blocker agent and because of a lack of time I decided not to put effort into this.