Difference between revisions of "Category:Advanced Targeting Strategies"

From Robowiki
Jump to navigation Jump to search
m (re-specify as any learning gun, maybe rename later)
m (more cleanup)
 
Line 1: Line 1:
 
Targeting strategies that collect data and learn how an enemy moves. There are a wide range of algorithms that fit this category, but most include one or more of the following:  
 
Targeting strategies that collect data and learn how an enemy moves. There are a wide range of algorithms that fit this category, but most include one or more of the following:  
  
* [[Waves]], a system for collection of [[GuessFactor]]s or [[Bearing Offsets]] that hit the enemy.
+
* [[Waves]] - A mechanic for collecting [[GuessFactor]]s or [[Bearing Offset|bearing offset]] that would hit the enemy.
* [[Pattern Matching]], a system that logs individual enemy movements, frame by frame, for analysis of past movements and projection of future movements.
+
* [[Pattern Matching]] - A system that logs individual enemy movements, frame by frame, for analysis of past movements and projection of future movements.
* [[Dynamic Clustering]], where the current situation is compared to a log of many or all past situations to use as reference for the current situation.
+
* [[Visit Count Stats]] - A [[wikipedia:histogram|histogram]] of the frequency with which the enemy visits different firing angles.
 +
* [[Dynamic Clustering]] - The current situation is compared to a log of past situations and the most similar are used to determine a firing angle.
  
 
[[Category:Targeting|Advanced Targeting Strategies]]
 
[[Category:Targeting|Advanced Targeting Strategies]]

Latest revision as of 20:15, 8 September 2011

Targeting strategies that collect data and learn how an enemy moves. There are a wide range of algorithms that fit this category, but most include one or more of the following:

  • Waves - A mechanic for collecting GuessFactors or bearing offset that would hit the enemy.
  • Pattern Matching - A system that logs individual enemy movements, frame by frame, for analysis of past movements and projection of future movements.
  • Visit Count Stats - A histogram of the frequency with which the enemy visits different firing angles.
  • Dynamic Clustering - The current situation is compared to a log of past situations and the most similar are used to determine a firing angle.