I left robocoding a long time ago after seeing bin-based statistical algorithms owning the rumble, which I never liked. Hard-coded segmentation felt too artificial. More recently, after seeing dynamic clustering owning the rumble, which is a much cooler algorithm, and finally understanding what wave surfing is all about, it brought me back.
- mn.Combat 1.x - The first bot I made a long time ago. Virtual gun array of simple targeters (head-on, linear, circular, random) and pattern matching. Anti-gravity movement.
- mn.WarMachine - Failed attempt at a geometric wall avoidance and perpendicular movement (which I though was the same thing as wave surfing at the time).
- mn.Combat 2.x - Complete code rewrite, but kept the name. Dynamic clustering everywhere I could think of. Anti-gravity movement with wall avoidance/precise prediction.
- Slowly improve mn.Combat so it goes as high as possible in all league rankings (except Twin Duel).
- Help improve the RoboRumble infrastructure.
- Clog the participants list with weak sample bots.
- Complain about the APS rating system being too susceptible to king-maker scenarios.
- Reach world peace - Sounds silly but by understanding the mechanics of conflict and competition one can learn how to prevent them. And games are excellent environments for that.