Difference between revisions of "NUSBot"
| (One intermediate revision by the same user not shown) | |||
| Line 13: | Line 13: | ||
|externalDownload= | |externalDownload= | ||
|language=C++ | |language=C++ | ||
| − | | | + | |googleCode=https://code.google.com/p/nus-bot/ |
|site= | |site= | ||
|tlstream= | |tlstream= | ||
| Line 29: | Line 29: | ||
|featured= | |featured= | ||
}} | }} | ||
| + | |||
| + | This Bot attempts to deepen spatial and temporal awareness, for tactical and strategic reasoning. | ||
| + | This NUS-Bot is designed with a military operation manual of 'Intelligence Preparation of the Battlefield' (IPB), for iterative spatial (map) and temporal reasoning. | ||
| + | |||
| + | It integrates IPB into an Multi-Agent system, with different agents acting together to intelligently locate chokepoints that have highest strength of enemy/friendly forces. This is done by using abstracted Influence Map being referred to by Intelligent Transport Agent. The bot’s algorithm is able to determine strategic chokepoints along the main path of both players' base locations, triggering dropping particular Transport (with Units loaded) on certain critical chokepoint(s). With this strategy, NUS-Bot can create a greater advantage against its opponent, as most current bots only do stereotyped-travel along single main path once the enemy base is located. Thus it’s wise to deploy units for ambush and guerrilla tactics to surprise the enemy. Next release of NUS-Bot will be further optimized to operate on island maps, for which other bots can’t handle currently. | ||
| + | |||
| + | This NUS-Bot is an experimental research based on UAlbertaBot 2013 version. | ||
| + | |||
| + | Bot debut is in August 2014, for AIIDE 2014, IEEE CIG 2014. | ||
Latest revision as of 16:05, 23 July 2015
Singapore 
This Bot attempts to deepen spatial and temporal awareness, for tactical and strategic reasoning. This NUS-Bot is designed with a military operation manual of 'Intelligence Preparation of the Battlefield' (IPB), for iterative spatial (map) and temporal reasoning.
It integrates IPB into an Multi-Agent system, with different agents acting together to intelligently locate chokepoints that have highest strength of enemy/friendly forces. This is done by using abstracted Influence Map being referred to by Intelligent Transport Agent. The bot’s algorithm is able to determine strategic chokepoints along the main path of both players' base locations, triggering dropping particular Transport (with Units loaded) on certain critical chokepoint(s). With this strategy, NUS-Bot can create a greater advantage against its opponent, as most current bots only do stereotyped-travel along single main path once the enemy base is located. Thus it’s wise to deploy units for ambush and guerrilla tactics to surprise the enemy. Next release of NUS-Bot will be further optimized to operate on island maps, for which other bots can’t handle currently.
This NUS-Bot is an experimental research based on UAlbertaBot 2013 version.
Bot debut is in August 2014, for AIIDE 2014, IEEE CIG 2014.
