Difference between revisions of "Tomcat"

From Robowiki
Jump to navigation Jump to search
m (update)
(Add navbox)
 
(24 intermediate revisions by 2 users not shown)
Line 3: Line 3:
 
| name            = Tomcat
 
| name            = Tomcat
 
| altimage        =  
 
| altimage        =  
| image          =  
+
| image          = Tomcat_logo.jpg
| imagesize      =
+
| caption        = [[:Image:Tomcat_logo.jpg|Full Size]]
| caption        =  
 
 
| author          = [[User:Jdev]]
 
| author          = [[User:Jdev]]
 
| extends        = [[AdvancedRobot]]
 
| extends        = [[AdvancedRobot]]
| targeting      = [[RS]]-[[PIF]]
+
| targeting      = [[Dynamic_Clustering|kNN]]-[[PIF]]
 
| movement        = [[RS]]-[[Wave Surfing]]
 
| movement        = [[RS]]-[[Wave Surfing]]
 
| released        = August 5th, 2011
 
| released        = August 5th, 2011
| best_rating    = 83.04 APS (18nd)
+
| best_rating    = 88.14 APS (3rd)
| rating          = 83.04 APS (18nd)
+
| rating          = 88.27 APS (4rd)
| current_version = 3.33
+
| current_version = 3.68
 
| license        = [http://creativecommons.org/licenses/by-nd/3.0/deed.en CC BY-ND 3.0]
 
| license        = [http://creativecommons.org/licenses/by-nd/3.0/deed.en CC BY-ND 3.0]
| download_link  = http://jdevs-robocode.googlecode.com/files/lxx.Tomcat_3.33.jar
+
| download_link  = https://github.com/aleksey-zhidkov/Tomcat/raw/develop/builds/lxx.Tomcat_3.68.jar
 
| isOneOnOne      = true
 
| isOneOnOne      = true
 
| isMelee        = false
 
| isMelee        = false
 
| isOpenSource    = true
 
| isOpenSource    = true
 +
}}
 +
{{Navbox small
 +
| title        = Sub-pages
 +
| page1        = Version History
 
}}
 
}}
  
Line 25: Line 28:
  
 
==== What's special about it? ====
 
==== What's special about it? ====
* As i know it's only one bot, which use range search for data analyse
+
* It's the best Russian bot at now
* it's the best Russian bot at release moment
 
 
* It's my first released bot
 
* It's my first released bot
 
* It was under development without releases more than 2 years
 
* It was under development without releases more than 2 years
 
* It has only one gun
 
* It has only one gun
 
* "g" and "m" keys enables gun and movement debug graphics respectively
 
* "g" and "m" keys enables gun and movement debug graphics respectively
* Tomcat use unique buffers management system for movement
+
* Tomcat use unique adaptive logs management system in movement
 +
* Now, maybe temporary, i stop active robocoding at all, and in rare times, when i return to it, i develop [[Emerald]]
  
 
==== How competitive is it? ====
 
==== How competitive is it? ====
Fairly well, Tomcat is in [[The2000Club|The 2000 Club]] and in Top 20 now
+
Very, Tomcat is in [[The_2100_Club|The 2100 Club]] and one of top duelists at 19 of september 2014
  
 
==== How does it [[:Category:Movement|move]]? ====
 
==== How does it [[:Category:Movement|move]]? ====
Tomcat has two sets of log - visits and hits. Each set - it's all possible subsets of {lateralVelocity, acceleration, distanceBetween and distanceToForwardWall}. For each log Tomcat tracks it's average "hit" and "miss" rates for last 3, 11, 100 bullets/waves. Each log can produce several (now 5) possible bearing offsets.<br/>
+
===== Data analysis =====
 +
Tomcat has two sets of log - visits and hits. Each set - it's all possible subsets of {lateralVelocity, acceleration, distanceBetween, distanceToForwardWall, distLast10Ticks}. For each log Tomcat tracks it's average "hit" and "miss" rates for last 9, 45, 5000 bullets/waves and enemy hit rate, when this log was used. Each log can produce several (now 5) possible bearing offsets.<br/>
 
When enemy bullet hit Tomcat or his bullet, Tomcat calculate "hit" rate for each log - how many predicted bullets contained by interval with robot width and center in real bearing offset.<br/>
 
When enemy bullet hit Tomcat or his bullet, Tomcat calculate "hit" rate for each log - how many predicted bullets contained by interval with robot width and center in real bearing offset.<br/>
 
When enemy bullet missed, Tomcat calculate "miss" rate for each log - how many predicted bullets contained by interval with robot width and center in current bearing offset.<br/>
 
When enemy bullet missed, Tomcat calculate "miss" rate for each log - how many predicted bullets contained by interval with robot width and center in current bearing offset.<br/>
When it's time to predict enemy's bullet bearing offset, Tomcat selects by 2 best logs in terms of "hit rate" - "miss rate" for last 3, 11 and 100 bullets and tries to dodge bullets produced by this 6 logs (there're may be only 2 logs if they is the best in all categories)
+
When it's time to predict enemy's bullet bearing offset, Tomcat selects by 3 best logs in terms of "hit rate" - "miss rate" for last 9, 45 and 5000 bullets/ and 3 best logs in terms of enemy's hit rate. Then Tomcat tries to dodge bullets produced by this 12 logs (there're may be only 3 logs if they are the best in all categories)
 +
 
 +
===== Movement mechanics =====
 +
Tomcat has pretty unique movement mechanics, which is hybrid between True Surfing and GoTo Surfing. Tomcat's reselect destination point only after few events: first bullet changed (passed, hitted or intercepted), bullet data updated (mainly because more late bullet has been intercepted), enemy change velocity sign and Tomcat's distance to destination point is less than stop distance + 8.<br>
 +
For reselect orbit direction Tomcat generate two destination points, which encourage him to move in clockwise and counter clockwise movement at max speed. These points are selected on circle with center in bullet fire position and radius = (distance between bullet fire position and Tomcat) * 10. Then Tomcat calculates precise route in both direction and danger of each point of routes. Then the most safe point are selected as destination point. Tomcat do not directly go to this point, but each tick do decision in wich orbit direction he must move to be at same bearing offset as this point. But really, because reselecting orbit direction when distance to destination point is less than stop distance + 8, this mechanics works like GoTo.
  
 
==== How does it [[:Category:Targeting|fire]]? ====
 
==== How does it [[:Category:Targeting|fire]]? ====
 
===== Data analysis =====
 
===== Data analysis =====
Tomcat keeps log of turns organized as linked list with several range search trees with different segmentation attributes for search. When it's time to fire, Tomcats find out set of similar situations in past, using every tree.  
+
Tomcat keeps log of turns organized as linked list with several kD-trees with different segmentation attributes for search. When it's time to fire, Tomcats find out set of similar situations in past, using every tree.  
 
===== Fire angle reconstruction =====
 
===== Fire angle reconstruction =====
 
See [[PIF/Gradient PIF]]
 
See [[PIF/Gradient PIF]]
Line 59: Line 67:
  
 
==== Can I use your code? ====
 
==== Can I use your code? ====
Yes, but you can only read code. See [http://creativecommons.org/licenses/by-nd/3.0/deed.en CC BY-ND 3.0]
+
You can use any code published on this wiki. But code avaliable from Tomcat's jars, you can only read. See [http://creativecommons.org/licenses/by-nd/3.0/deed.en CC BY-ND 3.0] for license of not published code.
  
 
==== What's next for your robot? ====
 
==== What's next for your robot? ====
* new movement mechanics
+
* Relax and be happy:)
* updating enemy fire angle prediction data for bullets in air, when new data income
 
* movement & targeting tweaks
 
* incorporate data about safe areas, recieved from waves intersections
 
* anti ram stuff, if new movement mechanics isn't help
 
* anti mirroring stuff
 
* melee strategy
 
* teams strategy
 
* optimisation
 
* try to add movement's logs selection system to gun
 
  
 
==== Does it have any [[WhiteWhale]]s? ====
 
==== Does it have any [[WhiteWhale]]s? ====
* The problem bots is mirror bots, ram bots and [[Pytko]]:)
+
In past Tomcat could beat any bot in RR, but now days [[DrussGT]] is big problem and [[Diamond]] wins with little advantage
* [[DrussGT]] - main insulter of Tomcat now:)
 
  
 
==== What other robot(s) is it based on? ====
 
==== What other robot(s) is it based on? ====
Set of my unreleased bots: Jdev, UltraMarine, Primarch; <br>And [[User:Rednaxela/FastTrig]] class and exact Non-Iterative solution for [[LinearTargeting]] from wiki
+
Set of my unreleased bots: Jdev, UltraMarine, Primarch; <br>[[User:Rednaxela/FastTrig]] class, [[User:Rednaxela/kD-Tree]] class and exact Non-Iterative solution for [[LinearTargeting]] from wiki
  
 
[[Category:Bots|Tomcat]]
 
[[Category:Bots|Tomcat]]
 
[[Category:1-vs-1 Bots|Tomcat]]
 
[[Category:1-vs-1 Bots|Tomcat]]
 
[[Category:MegaBots|Tomcat]]
 
[[Category:MegaBots|Tomcat]]

Latest revision as of 20:25, 9 August 2017

Tomcat
Tomcat logo.jpg
Full Size
Author(s) User:Jdev
Extends AdvancedRobot
Targeting kNN-PIF
Movement RS-Wave Surfing
Released August 5th, 2011
Best Rating 88.14 APS (3rd)
Current Rating 88.27 APS (4rd)
Current Version 3.68
Code License CC BY-ND 3.0
Download
Sub-pages:
Version History

Most Recent Updates

Check out the /Version History

What's special about it?

  • It's the best Russian bot at now
  • It's my first released bot
  • It was under development without releases more than 2 years
  • It has only one gun
  • "g" and "m" keys enables gun and movement debug graphics respectively
  • Tomcat use unique adaptive logs management system in movement
  • Now, maybe temporary, i stop active robocoding at all, and in rare times, when i return to it, i develop Emerald

How competitive is it?

Very, Tomcat is in The 2100 Club and one of top duelists at 19 of september 2014

How does it move?

Data analysis

Tomcat has two sets of log - visits and hits. Each set - it's all possible subsets of {lateralVelocity, acceleration, distanceBetween, distanceToForwardWall, distLast10Ticks}. For each log Tomcat tracks it's average "hit" and "miss" rates for last 9, 45, 5000 bullets/waves and enemy hit rate, when this log was used. Each log can produce several (now 5) possible bearing offsets.
When enemy bullet hit Tomcat or his bullet, Tomcat calculate "hit" rate for each log - how many predicted bullets contained by interval with robot width and center in real bearing offset.
When enemy bullet missed, Tomcat calculate "miss" rate for each log - how many predicted bullets contained by interval with robot width and center in current bearing offset.
When it's time to predict enemy's bullet bearing offset, Tomcat selects by 3 best logs in terms of "hit rate" - "miss rate" for last 9, 45 and 5000 bullets/ and 3 best logs in terms of enemy's hit rate. Then Tomcat tries to dodge bullets produced by this 12 logs (there're may be only 3 logs if they are the best in all categories)

Movement mechanics

Tomcat has pretty unique movement mechanics, which is hybrid between True Surfing and GoTo Surfing. Tomcat's reselect destination point only after few events: first bullet changed (passed, hitted or intercepted), bullet data updated (mainly because more late bullet has been intercepted), enemy change velocity sign and Tomcat's distance to destination point is less than stop distance + 8.
For reselect orbit direction Tomcat generate two destination points, which encourage him to move in clockwise and counter clockwise movement at max speed. These points are selected on circle with center in bullet fire position and radius = (distance between bullet fire position and Tomcat) * 10. Then Tomcat calculates precise route in both direction and danger of each point of routes. Then the most safe point are selected as destination point. Tomcat do not directly go to this point, but each tick do decision in wich orbit direction he must move to be at same bearing offset as this point. But really, because reselecting orbit direction when distance to destination point is less than stop distance + 8, this mechanics works like GoTo.

How does it fire?

Data analysis

Tomcat keeps log of turns organized as linked list with several kD-trees with different segmentation attributes for search. When it's time to fire, Tomcats find out set of similar situations in past, using every tree.

Fire angle reconstruction

See PIF/Gradient PIF

Multiple choice

When enemy's possible future positions is finded out, Tomcat calulates interval of bearing offsets (BO), which hits enemy for each future position. Then, Tomcat builds histogram of BOs dangers and selects random BO from set of BOs with maximum danger value.

Implementation features

Tomcat's PIF algorithm is position based - it's calculates enemy displacement vector between two turns in past, which is applied to enemy's current state to calculate enemy's future position. This algorthm allows me to avoid accumulating rounding errors

What does it save between rounds and matches?

Between rounds it saves targeting and movement logs and other stuff (enemy prefered disance, enemy targeting profile, hit rates etc.), between matches nothing

Where did you get the name?

From his colors:)

Can I use your code?

You can use any code published on this wiki. But code avaliable from Tomcat's jars, you can only read. See CC BY-ND 3.0 for license of not published code.

What's next for your robot?

  • Relax and be happy:)

Does it have any WhiteWhales?

In past Tomcat could beat any bot in RR, but now days DrussGT is big problem and Diamond wins with little advantage

What other robot(s) is it based on?

Set of my unreleased bots: Jdev, UltraMarine, Primarch;
User:Rednaxela/FastTrig class, User:Rednaxela/kD-Tree class and exact Non-Iterative solution for LinearTargeting from wiki