Difference between revisions of "Combat"
Jump to navigation
Jump to search
m (Removed []) |
m (Category) |
||
Line 57: | Line 57: | ||
; What other robot(s) is it based on? | ; What other robot(s) is it based on? | ||
The code is 100% original. But I took most of the ideas from RoboWiki. | The code is 100% original. But I took most of the ideas from RoboWiki. | ||
+ | |||
+ | [[Category:Bots|Combat]] |
Revision as of 18:24, 13 August 2011
Background Information
- What's special about it?
It mixes Old Skool techniques I like with a few "modern" techniques with better than expected results.
Strategy
- How does it move?
- Old Skool anti-gravity movement boosted with precise prediction/wall avoidance.
- It delays wall avoidance until the last possible tick. Not as strong as wall smoothing in 1v1, but it is cool to see the bot making aggressive turns when near walls.
- How does it fire?
- Guess factor targeting with dynamic clustering segmented data. But I stripped that direction attribute away. 1 is always clockwise and -1 is always counter-clockwise. Direction is accounted indirectly as a dynamic clustering signed lateral velocity classifier. Feels less artificial that way and confuses some wave surfing implementations.
- How does it dodge bullets?
- Old Skool shrapnel dodging. Anti-gravity enemy virtual bullets generated with dynamic clustering segmented data. Works well in melee/team and it is cool to watch. Working better than expected in 1v1.
- How does the melee strategy differ from one-on-one strategy?
- It doesn´t, but it will some day.
- How does it select a target to attack/avoid in melee ?
- What does it save between rounds and matches?
- Between matches: Nothing.
- Between rounds: Wave data.
Additional Information
- Where did you get the name?
- From the Atari 2600 game Combat.
- Can I use your code?
- Not open-source.
- What's next for your robot?
- Cool algorithms
- A k-d tree to optimize dynamic clustering. It is skipping turns to the point of being disabled in teamrumble.
- Energy management with dynamic clustering segmented data for hit rate estimation. Should be cool to watch.
- Precise prediction/max escape angle calculation. Shooting walls is not cool to watch.
- Improved precise prediction/wall avoidance. It is shaking (turning left and right) when moving near walls. Not cool to watch.
- Precise prediction/anti-gravity movement to calculate when to stop. It is vibrating (moving ahead and back) instead of stopping. Not cool to watch.
- Bomb sheltering evaluation to erase enemy virtual bullets.
- Rating improvement
- Go to surfing for 1v1.
- Anti-surfer gun for 1v1.
- Curve flattening for 1v1.
- What other robot(s) is it based on?
The code is 100% original. But I took most of the ideas from RoboWiki.