Difference between revisions of "ScalarR"
m |
m (Minor update) |
||
| Line 18: | Line 18: | ||
; What's special about it? | ; What's special about it? | ||
| − | It's a melee bot, and shares same surfing algorithms in 1v1. | + | It's a melee bot, and shares same surfing algorithms in 1v1. More details described in [[ScalarR/General Surfing Framework]]. |
; How competitive is it? | ; How competitive is it? | ||
| Line 24: | Line 24: | ||
No.1 in melee ([http://web.archive.org/web/20190319084925/http://literumble.appspot.com/Rankings?game=meleerumble February 12, 2019]) | No.1 in melee ([http://web.archive.org/web/20190319084925/http://literumble.appspot.com/Rankings?game=meleerumble February 12, 2019]) | ||
| − | No. | + | No.2 in 1v1 (No.1 on [http://web.archive.org/web/20210628051845/https://literumble.appspot.com/Rankings?game=roborumble June 28, 2021]) |
== Strategy == | == Strategy == | ||
| Line 50: | Line 50: | ||
; What's next for your robot? | ; What's next for your robot? | ||
| − | * | + | * [[GoTo Surfing]] looks promising, but requires different danger estimation than [[True Surfing]]. |
| − | |||
* Tune gun / movement with more computation power. | * Tune gun / movement with more computation power. | ||
| − | * | + | * Rework energy management, probably trying something [[wikipedia:Reinforcement Learning|RL]]. |
; What other robot(s) is it based on? | ; What other robot(s) is it based on? | ||
Revision as of 07:17, 23 July 2021
- Sub-pages:
- Version History - Some Ideas
| ScalarR | |
| Author(s) | Xor |
| Extends | AdvancedRobot |
| Targeting | GuessFactor Targeting |
| Movement | Wave Surfing |
Information on this page may get outdated. For recent activities, see Version History.
Background Information
- What's special about it?
It's a melee bot, and shares same surfing algorithms in 1v1. More details described in ScalarR/General Surfing Framework.
- How competitive is it?
No.1 in melee (February 12, 2019)
No.2 in 1v1 (No.1 on June 28, 2021)
Strategy
- How does it move?
True Surfing in melee and 1v1, with different surfing paths generated. Danger computation considers Bullet Shadow, KNN and a simulation of simple guns (to encounter its weakness agains weak bots ;).
- How does it fire?
KNN/Play It Forward in melee, KNN/GuessFactor Targeting in 1v1. Energy Management is quite conservative that optimizes survival as main goal. The weights of 1v1 main KNN gun is tuned with Genetic Algorithms inspired by Skilgannon, by directly optimizing hit rate. The 1v1 anti-surfing gun is not tuned yet.
- What does it save between rounds and matches?
Everything between rounds, nothing between matches.
Additional Information
- Where did you get the name?
Scalar from Scalar Replacement, an optimization technique used in JVM to reduce GC pressure. Letter R is coined.
- What's next for your robot?
- GoTo Surfing looks promising, but requires different danger estimation than True Surfing.
- Tune gun / movement with more computation power.
- Rework energy management, probably trying something RL.
- What other robot(s) is it based on?
Directly and indirectly influenced by everything mentioned on this wiki.
- Experience learnt from ScalarN, ScalarBot and SimpleBot
- Understanding DrussGT. Thanks Skilgannon for sharing many details about how to build a competitive bot.
- Innovations since 2005
- Diamond’s code is very neat, I learned a lot but still get messy code.
- Understanding BeepBoop. Thanks Kev for sharing the idea about using gradients, as well as showing how import correct bullet shadow is.
- to be listed...
