Fractals2019: Combinatorial Optimisation with Dynamic Constraint Annealing
This work addresses tactical optimization in multi-agent robotic soccer, representing an incremental improvement over existing methods like Gliders2d.
The paper tackled the problem of optimizing collective behaviors in RoboCup Soccer 2D by applying combinatorial optimization with Dynamic Constraint Annealing, resulting in the Fractals2019 team becoming the RoboCup-2019 champion.
Fractals2019 started as a new experimental entry in the RoboCup Soccer 2D Simulation League, based on Gliders2d code base, and advanced to become a RoboCup-2019 champion. We employ combinatorial optimisation methods, within the framework of Guided Self-Organisation, with the search guided by local constraints. We present examples of several tactical tasks based on the Gliders2d code (version v2), including the search for an optimal assignment of heterogeneous player types, as well as blocking behaviours, offside trap, and attacking formations. We propose a new method, Dynamic Constraint Annealing, for solving dynamic constraint satisfaction problems, and apply it to optimise thermodynamic potential of collective behaviours, under dynamically induced constraints.