microPhantom: Playing microRTS under uncertainty and chaos
This work addresses the problem of AI decision-making in real-time strategy games under uncertainty for researchers and developers, representing an incremental improvement over previous bots.
The authors tackled decision-making under uncertainty in the microRTS game by developing microPhantom, a bot that uses Constraint Programming and decision theory to solve the Unit Production Problem, resulting in a significant improvement in win rate against the second-best bot in partially observable conditions and showing resilience with minimal efficiency loss in chaotic environments.
This competition paper presents microPhantom, a bot playing microRTS and participating in the 2020 microRTS AI competition. microPhantom is based on our previous bot POAdaptive which won the partially observable track of the 2018 and 2019 microRTS AI competitions. In this paper, we focus on decision-making under uncertainty, by tackling the Unit Production Problem with a method based on a combination of Constraint Programming and decision theory. We show that using our method to decide which units to train improves significantly the win rate against the second-best microRTS bot from the partially observable track. We also show that our method is resilient in chaotic environments, with a very small loss of efficiency only. To allow replicability and to facilitate further research, the source code of microPhantom is available, as well as the Constraint Programming toolkit it uses.