team-remember-to-hydrate / battlecode23-team-remember-to-hydrate

GNU Affero General Public License v3.0
1 stars 0 forks source link

Strategy: Dynamic Targeting Priorities #13

Open mklapp2 opened 1 year ago

mklapp2 commented 1 year ago

Goal: Implement MVP of a targeting strategy that can be switched between modes

Description: We can all imagine that it is better if launchers (and carriers) pick their targets non-randomly. We could hard-code a priority order that remains static throughout a match. However, it might be better if we can pre-code targeting priority 'modes' that can be switched between mid-match (either at a team level, or maybe a 'group' level if we implement group assignments.

The deliverable of this issue would be a methodized way to switch between pre-configured targeting prioritization modes.

Some general heuristics that might be prioritized between:

A separate task would be adjusting bot logic regarding 'locking on' to a target and how that could make targeting more efficient (I believe you can use less bytecode if you know your target's ID and know you just want their location to attack them again if they are in sight range).

Odyhibit commented 1 year ago

Since each turn the targeting would probably be recalculated each turn for a minimum viable product this could be as simple as creating conditions that define the targeting mode, then a switch statement implementing them.