Difference between revisions of "Oculus"

From Robowiki
Jump to navigation Jump to search
(Made myself a promise)
 
(3 intermediate revisions by the same user not shown)
Line 12: Line 12:
 
| extends        = [[AdvancedRobot]]
 
| extends        = [[AdvancedRobot]]
 
| targeting      = [[Neural Targeting]] (GF)
 
| targeting      = [[Neural Targeting]] (GF)
| movement        = [[Neural Movement]] (WS-GF)
+
| movement        = Stochastic [[Neural Movement]] (WS-GF)
 
| released        = 21.11.2017
 
| released        = 21.11.2017
 
| rating          = x
 
| rating          = x
| current_version = 1.3.6
+
| current_version = 1.5
| download_link  = https://www.dropbox.com/s/jtzfxixiwlmqqfc/dsekercioglu.neural.Oculus_1.3.6.jar?dl=1
+
| download_link  = https://www.dropbox.com/s/djbubltrfj3pzv5/dsekercioglu.neural.Oculus_1.5.jar?dl=0
 
| isOneOnOne      = true
 
| isOneOnOne      = true
 
| isMelee        = false
 
| isMelee        = false
Line 34: Line 34:
  
 
; What's special about it?
 
; What's special about it?
: ''It is my first neural robot which isn't related to [[Gaff]]''
+
: ''It is one of the only bullet shielders like all my other bots.''
 +
: ''It uses a stochastic neural network view.''
  
 
; ''Great, I want to try it. Where can I download it.''
 
; ''Great, I want to try it. Where can I download it.''
: ''https://www.dropbox.com/s/jtzfxixiwlmqqfc/dsekercioglu.neural.Oculus_1.3.6.jar?dl=1''
+
: ''https://www.dropbox.com/s/djbubltrfj3pzv5/dsekercioglu.neural.Oculus_1.5.jar?dl=1''
  
 
; ''How competitive is it?''
 
; ''How competitive is it?''
: ''Not very right now.''
+
: ''We will see.''
  
 
;Credits
 
;Credits
 
: Credits should go to:
 
: Credits should go to:
 
:[[Rozu]]: predictPosition from [[Apollon]].
 
:[[Rozu]]: predictPosition from [[Apollon]].
 +
:[[Chase-san]] the intercept method which I use for [[PPMEA]].
  
 
== Strategy ==
 
== Strategy ==
Line 55: Line 57:
  
 
; How does it [[Dodging Bullets|dodge bullets]]?
 
; How does it [[Dodging Bullets|dodge bullets]]?
: ''Trying to predict the dangers of [[GF]]s with a Neural Network.''
+
: ''It uses true surfing.''
  
 
; How does the [[melee]] strategy differ from [[One-on-one]]  strategy?
 
; How does the [[melee]] strategy differ from [[One-on-one]]  strategy?
: ''Nothing.''
+
: ''It doesn't support melee''
  
 
; How does it select a target to attack/avoid in [[melee]]?
 
; How does it select a target to attack/avoid in [[melee]]?
: ''The first enemy it sees.''
+
: ''I doesn't support melee.''
  
 
; What does it save between rounds and matches?
 
; What does it save between rounds and matches?
: ''[[NN]] weights.''
+
: ''[[NN]] weights and hit rates.''
  
 
== Additional Information ==
 
== Additional Information ==
Line 72: Line 74:
  
 
; Why did you make this bot?
 
; Why did you make this bot?
: ''A new type of pre-processing worked very well.''
+
: ''This is actually the third time I rewrite this robot. The reason is that I now have a better Neural Network framework.''
  
 
; Can I use your code?
 
; Can I use your code?
: ''Of course with a decompiler.''
+
: ''Java files are in the jar file.''
  
 
; What's next for your robot?
 
; What's next for your robot?
: ''Making the gun an RM killer.''
+
: ''Do the fourth rewrite if I get WF to top 10''
: ''Add BS Learning.''
+
: ''Tune everything''
: ''Bullet Shielding.''
 
  
 
; Does it have any [[White Whale]]s?
 
; Does it have any [[White Whale]]s?
: ''Every bot I make does really bad against [[Tron]].''
+
: ''We will see.''
  
 
; What other robot(s) is it based on?
 
; What other robot(s) is it based on?
: ''None''
+
: ''My standard robot base.''
  
 
[[Category:Templates|Bots]]
 
[[Category:Templates|Bots]]

Latest revision as of 20:46, 20 March 2019

Oculus Sub-pages:
Version History - Challenges


Oculus
Author(s) Dsekercioglu
Extends AdvancedRobot
Targeting Neural Targeting (GF)
Movement Stochastic Neural Movement (WS-GF)
Released 21.11.2017
Current Rating x
Current Version 1.5
Download

Background Information

Bot Name
Oculus
Author
Dsekercioglu
Extends
AdvancedRobot
What's special about it?
It is one of the only bullet shielders like all my other bots.
It uses a stochastic neural network view.
Great, I want to try it. Where can I download it.
https://www.dropbox.com/s/djbubltrfj3pzv5/dsekercioglu.neural.Oculus_1.5.jar?dl=1
How competitive is it?
We will see.
Credits
Credits should go to:
Rozu: predictPosition from Apollon.
Chase-san the intercept method which I use for PPMEA.

Strategy

How does it move?
True Surfing
How does it fire?
Trying to predict the probabilities of GFs with a Neural Network.
How does it dodge bullets?
It uses true surfing.
How does the melee strategy differ from One-on-one strategy?
It doesn't support melee
How does it select a target to attack/avoid in melee?
I doesn't support melee.
What does it save between rounds and matches?
NN weights and hit rates.

Additional Information

Where did you get the name?
Eye in latin.
Why did you make this bot?
This is actually the third time I rewrite this robot. The reason is that I now have a better Neural Network framework.
Can I use your code?
Java files are in the jar file.
What's next for your robot?
Do the fourth rewrite if I get WF to top 10
Tune everything
Does it have any White Whales?
We will see.
What other robot(s) is it based on?
My standard robot base.