Difference between revisions of "AgentSmith/Wolfmans Todo List"

Revision as of 23:12, 26 November 2013

Agent Smith Sub-pages:: AgentSmith - Version History - Challenge Results - Wolfmans Todo List

Goals - In order, 1v1

@@ Line 7: / Line 7: @@
 }}
-Short Term
+Goals - In order, 1v1
-* Get the gun to a reasonable standard - add more segmentation
-* Use targeting challenge results to tune gun
-* <strike>Weight normalised DC values for unbounded segments in a better manner (such as time-since values etc)</strike> - DONE - ''but performance decreased. Needs investigating''
-* <strike>Factor in distance from the best match when deciding what angle to shoot at based on our K nearest matches.</strike> - DONE
-* <strike>Store dynamic DC weights per bot between matches to give the targeting a head start even if we don't store any of the tree.</strike> - ''Dynamic weighting removed. Needs more thinking.''
-* Add in really quick and dirty random movement
-* Release into the rumble
-* Calculate MEA using [[Precise Prediction]] and [[Wall Smoothing]]
-Long Term
+* Beat my old robot Warlord
-* Multiple gun modes chosen dynamically based on match state
+* Get the crown of Best Robot in the UK currently held by Sandbox Lump AFAIK.
-* Anti-surfing DC gun mode
+* Get into the top 100
-* Wave surfing movement
+* Get into the top 10
-* Genetic algorithms to tune various constants in the bot
+* Get 1st position
-* Investigate storing small pre-populated data set for faster DC learning in a fresh match scenario