This is the implementation of an MCTS-based multi-player board game for robotic team composition optimization.
pandas and numpy
- Multi-tasking robotic team with the same robot type: The robot can handle all types of activities from A to H
python multi_tasking_team.py
--total_game <Number of games to play>
--player_num <Number of players of the same type>
--N <Number of simulations per round>
--C <Parameter for balancing utilization and exploration>
--scaffold_type <2x1|2x2|2x4|2x6|2x8|2x10>
For example, 1 game, 3 robots, and 2 story 2 span scaffold
python multi_tasking_team.py --total_game 1 --player_num 3 --N 50 --C 10 --scaffold_type 2x2
- Mixed robotic team with two different types of robots: Installation robots [C D F H] and transportation robots [A B E G]
python mixed_team.py
--total_game <Number of games to play>
--humanoid_num <Number of humanoid robots>
--robot_num <Number of general transportation robots>
--N <Number of simulations per round>
--C <Parameter for balancing utilization and exploration>
--scaffold_type <2x1|2x2|2x4|2x6|2x8|2x10>
For example, 1 game, 2 installation robots, 1 transportation robot, and 2 story 2 span scaffold
python mixed_team.py --total_game 1 --humanoid_num 2 --robot_num 1 --N 50 --C 10 --scaffold_type 2x2