Difference between revisions of "Diamond"
Jump to navigation
Jump to search
m |
(updates) |
||
| Line 4: | Line 4: | ||
| caption = [[:Image:Diamond_collage.png|Full Size]] | | caption = [[:Image:Diamond_collage.png|Full Size]] | ||
| extends = [[AdvancedRobot]] | | extends = [[AdvancedRobot]] | ||
| − | | targeting = [[Dynamic | + | | targeting = [[Dynamic Clustering]], |
| − | | movement = [[Minimum Risk Movement|Minimum | + | [[Displacement Vector]]s, [[GuessFactor]]s |
| − | | current_version = 1. | + | | movement = [[Minimum Risk Movement|Minimum Risk]], |
| + | [[Dynamic Clustering|DC]]/[[GuessFactor|GF]]-[[Wave Surfing]] | ||
| + | | current_version = 1.7.11 | ||
| license = [http://www.gzip.org/zlib/zlib_license.html zlib] | | license = [http://www.gzip.org/zlib/zlib_license.html zlib] | ||
| − | | download_link = http://www.dijitari.com/void/robocode/voidious.Diamond_1. | + | | download_link = http://www.dijitari.com/void/robocode/voidious.Diamond_1.7.11.jar |
| isOpenSource = yes | | isOpenSource = yes | ||
| isMelee = yes | | isMelee = yes | ||
| Line 23: | Line 25: | ||
; What's special about it? | ; What's special about it? | ||
: Oh, nothing, really. Well, since you asked... | : Oh, nothing, really. Well, since you asked... | ||
| − | :* | + | :* Took the [[Melee]] throne from [[Shadow]]. |
| − | :* | + | :* By far the top [[Dynamic Clustering|DC]] surfer. |
| − | :* | + | :* Pioneered using [[Bullet Shadow]]s in Wave Surfing. |
| − | :* | + | :* The Melee [[Wall Smoothing]] is quite graceful. |
| − | :* | + | :* Melee guns use [[Displacement Vector|displacement vectors]] and fire lots of waves from alternate sources. |
| − | :* | + | :* Pretty debugging graphics. |
| + | :* Two alternate color schemes! | ||
:* The source code is available under a [[wikipedia:Permissive free software licence|permissive license]] - have fun! (And try to learn something!) | :* The source code is available under a [[wikipedia:Permissive free software licence|permissive license]] - have fun! (And try to learn something!) | ||
; How competitive is it? | ; How competitive is it? | ||
| − | : Very | + | : Very! #2 in Melee right behind [[DemonicRage]], #2 in 1v1 behind [[DrussGT]]. |
== Strategy == | == Strategy == | ||
; How does it [[:Category:Movement|move]]? | ; How does it [[:Category:Movement|move]]? | ||
| − | : In melee, it uses a [[Minimum Risk Movement]], evaluating a bunch of points around itself and choosing the best | + | : In melee, it uses a [[Minimum Risk Movement]], evaluating a bunch of points around itself and choosing the best one. I'm always tweaking the parameters, but some of them are: |
| − | + | :* Stay far away from other bots. | |
| − | :* Stay far away from other bots | ||
:* Stay perpendicular to other bots. | :* Stay perpendicular to other bots. | ||
| − | :* Don't be the closest bot to any other bots (avoid being targeted). | + | :* Don't be the closest bot to any other bots (to avoid being targeted). |
| + | :* Weight risks by energy and bullet damage factors. | ||
:* Some randomizing based on past locations. | :* Some randomizing based on past locations. | ||
| − | : In 1v1, it uses [[Wave Surfing]] | + | |
| + | : In 1v1, it uses [[Wave Surfing]] with [[Dynamic Clustering]]. | ||
| + | :* [[Precise Prediction|Predict]]s where he would intersect the enemy [[wave]] if he orbits clockwise, counter-clockwise, or slams on the brakes. Calculate the danger for this position based on where we've detected enemy bullets. | ||
| + | :* For each predicted spot on the first wave, branch from that position and predict clockwise/counter-clockwise/stop on the second wave. Take the lowest danger. | ||
| + | :* Weight the wave dangers by time to impact and bullet power. | ||
| + | :* Multiply in a distancing factor. | ||
| + | :* Choose the safest of the 3 movement options. | ||
; How does it [[:Category:Targeting|fire]]? | ; How does it [[:Category:Targeting|fire]]? | ||
| − | : The gun uses Dynamic Clustering for data analysis - i.e., [http://en.wikipedia.org/wiki/K-nearest_neighbor_algorithm k-nearest neighbors] to find similar states, and [http://en.wikipedia.org/wiki/Kernel_density_estimation kernel density] among those states to settle on a firing angle. In Melee, it uses [[Displacement Vector|displacement vectors]] for recording and reconstructing firing angles - I | + | : The gun uses Dynamic Clustering for data analysis - i.e., [http://en.wikipedia.org/wiki/K-nearest_neighbor_algorithm k-nearest neighbors] to find similar states, and [http://en.wikipedia.org/wiki/Kernel_density_estimation kernel density] among those states to settle on a firing angle. |
| − | : I | + | |
| + | : In Melee, it uses [[Displacement Vector|displacement vectors]] for recording and reconstructing firing angles - I think that makes more sense than [[GuessFactor|GuessFactors]] since enemies aren't necessarily moving relative to you, and it's a lot simpler and more efficient than real [[Play It Forward]]. In 1v1, it uses GuessFactors with [[Maximum Escape Angle/Precise|precise MEA]]. It uses [[Waves/Precise Intersection|precise intersection]] and interpolates missed scans when necessary. | ||
| + | |||
| + | : I also experiment with gun stuff quite a bit with [[User:Voidious/WaveSim|WaveSim]]. My latest experiments involve using a [[RetroGirl/Gun]] for the first few shots of a 1v1 battle. | ||
; How does it select a target to attack/avoid in [[melee]]? | ; How does it select a target to attack/avoid in [[melee]]? | ||
| Line 53: | Line 65: | ||
; How does the [[Melee|melee]] strategy differ from [[One on One|one-on-one]] strategy? | ; How does the [[Melee|melee]] strategy differ from [[One on One|one-on-one]] strategy? | ||
| − | : In 1v1, it switches to a Wave Surfing movement and | + | : In 1v1, it switches to a Wave Surfing movement and to its 1v1 [[Virtual Guns]]. |
; What does it save between rounds and matches? | ; What does it save between rounds and matches? | ||
| − | : Nothing between matches. Between rounds, it saves all its targeting and movement data, which includes general info about each enemy, a bunch of [[Kd-Tree|kd-trees]], and maps of the trees' data points to displacement vectors and | + | : Nothing between matches. Between rounds, it saves all its targeting and movement data, which includes general info about each enemy, a bunch of [[Kd-Tree|kd-trees]], and maps of the trees' data points to displacement vectors and GuessFactors. |
== Additional Information == | == Additional Information == | ||
| Line 67: | Line 79: | ||
; What's next for your robot? | ; What's next for your robot? | ||
| − | : I'd love to take the 1v1 throne. A lot. =) | + | : I'd love to take the 1v1 throne. A lot. =) I'm also doing a huge refactor right now and adding unit tests, which is producing a steady stream of minor bug fixes and improvements, and hopefully some bigger ideas along the way. |
; Does it have any [[White Whale|White Whales]]? | ; Does it have any [[White Whale|White Whales]]? | ||
| Line 74: | Line 86: | ||
; What other robot(s) is it based on? | ; What other robot(s) is it based on? | ||
: I tried to start with a clean slate on this one. But I've taken plenty of ideas from the wiki and utility code from a few places. | : I tried to start with a clean slate on this one. But I've taken plenty of ideas from the wiki and utility code from a few places. | ||
| − | :* [[BrokenSword]] - General guidance on various things, especially the minimum risk movement | + | :* [[BrokenSword]] - General guidance on various things, especially the minimum risk movement. |
:* [[Dookious]] - Diamond's surfing algorithm was originally ported from Dookious, along with a variety of utility functions, including [[Precise Prediction|precise prediction]] code based on [[User:Albert|Albert]]'s [[FuturePosition]]. | :* [[Dookious]] - Diamond's surfing algorithm was originally ported from Dookious, along with a variety of utility functions, including [[Precise Prediction|precise prediction]] code based on [[User:Albert|Albert]]'s [[FuturePosition]]. | ||
| − | + | :* [[HawkOnFire]] - No code, but it deserves a nod for the wisdom [[User:rozu|rozu]] shared about minimum risk movement via this bot. | |
| − | :* [[HawkOnFire]] - | ||
:* [[User:Corbos|Corbos]], [[User:Simonton|Simonton]] - Corbos for the idea to use a kd-tree to speed up our DC guns, Simonton for the bucket variant. | :* [[User:Corbos|Corbos]], [[User:Simonton|Simonton]] - Corbos for the idea to use a kd-tree to speed up our DC guns, Simonton for the bucket variant. | ||
:* [[Shadow]] - Pioneered the [[Shadow/Melee Gun]], which is an amazing advancement in Melee targeting. | :* [[Shadow]] - Pioneered the [[Shadow/Melee Gun]], which is an amazing advancement in Melee targeting. | ||
| − | :* [[User:Rednaxela|Rednaxela]] - | + | :* [[User:Rednaxela|Rednaxela]] - Introduced us to [[Waves/Precise Intersection|precise intersection]] and [[Gun Heat Waves]], and I also use his lightning fast [[User:Rednaxela/kD-Tree|kd-tree]]. |
[[Category:MegaBots]] | [[Category:MegaBots]] | ||
Revision as of 17:53, 8 June 2012
| Diamond | |
 Full Size | |
| Author(s) | Voidious |
| Extends | AdvancedRobot |
| Targeting | Dynamic Clustering, |
| Movement | Minimum Risk, |
| Current Version | 1.7.11 |
| Code License | zlib |
| Download | |
- Sub-pages:
- Version History - DiamondHawk
Background Information
- What's special about it?
- Oh, nothing, really. Well, since you asked...
- Took the Melee throne from Shadow.
- By far the top DC surfer.
- Pioneered using Bullet Shadows in Wave Surfing.
- The Melee Wall Smoothing is quite graceful.
- Melee guns use displacement vectors and fire lots of waves from alternate sources.
- Pretty debugging graphics.
- Two alternate color schemes!
- The source code is available under a permissive license - have fun! (And try to learn something!)
- How competitive is it?
- Very! #2 in Melee right behind DemonicRage, #2 in 1v1 behind DrussGT.
Strategy
- How does it move?
- In melee, it uses a Minimum Risk Movement, evaluating a bunch of points around itself and choosing the best one. I'm always tweaking the parameters, but some of them are:
- Stay far away from other bots.
- Stay perpendicular to other bots.
- Don't be the closest bot to any other bots (to avoid being targeted).
- Weight risks by energy and bullet damage factors.
- Some randomizing based on past locations.
- In 1v1, it uses Wave Surfing with Dynamic Clustering.
- Predicts where he would intersect the enemy wave if he orbits clockwise, counter-clockwise, or slams on the brakes. Calculate the danger for this position based on where we've detected enemy bullets.
- For each predicted spot on the first wave, branch from that position and predict clockwise/counter-clockwise/stop on the second wave. Take the lowest danger.
- Weight the wave dangers by time to impact and bullet power.
- Multiply in a distancing factor.
- Choose the safest of the 3 movement options.
- How does it fire?
- The gun uses Dynamic Clustering for data analysis - i.e., k-nearest neighbors to find similar states, and kernel density among those states to settle on a firing angle.
- In Melee, it uses displacement vectors for recording and reconstructing firing angles - I think that makes more sense than GuessFactors since enemies aren't necessarily moving relative to you, and it's a lot simpler and more efficient than real Play It Forward. In 1v1, it uses GuessFactors with precise MEA. It uses precise intersection and interpolates missed scans when necessary.
- I also experiment with gun stuff quite a bit with WaveSim. My latest experiments involve using a RetroGirl/Gun for the first few shots of a 1v1 battle.
- How does it select a target to attack/avoid in melee?
- As of 1.30, it no longer does. It aims at everyone on the battle field at once and tries to hit as many enemies as possible. (Big thanks to Shadow/Melee Gun for the excellent idea.)
- How does the melee strategy differ from one-on-one strategy?
- In 1v1, it switches to a Wave Surfing movement and to its 1v1 Virtual Guns.
- What does it save between rounds and matches?
- Nothing between matches. Between rounds, it saves all its targeting and movement data, which includes general info about each enemy, a bunch of kd-trees, and maps of the trees' data points to displacement vectors and GuessFactors.
Additional Information
- Where did you get the name?
- Diamond is the ex-superhero name of the main character in Powers, my favorite comic book series. I like to name all my bots after warriors of some sort, but I'd be lying if I said I didn't just think it sounds cool.
- Can I use your code?
- Yes, it's released under the zlib license.
- What's next for your robot?
- I'd love to take the 1v1 throne. A lot. =) I'm also doing a huge refactor right now and adding unit tests, which is producing a steady stream of minor bug fixes and improvements, and hopefully some bigger ideas along the way.
- Does it have any White Whales?
- DrussGT
- What other robot(s) is it based on?
- I tried to start with a clean slate on this one. But I've taken plenty of ideas from the wiki and utility code from a few places.
- BrokenSword - General guidance on various things, especially the minimum risk movement.
- Dookious - Diamond's surfing algorithm was originally ported from Dookious, along with a variety of utility functions, including precise prediction code based on Albert's FuturePosition.
- HawkOnFire - No code, but it deserves a nod for the wisdom rozu shared about minimum risk movement via this bot.
- Corbos, Simonton - Corbos for the idea to use a kd-tree to speed up our DC guns, Simonton for the bucket variant.
- Shadow - Pioneered the Shadow/Melee Gun, which is an amazing advancement in Melee targeting.
- Rednaxela - Introduced us to precise intersection and Gun Heat Waves, and I also use his lightning fast kd-tree.
