Evolution of team composition in multi-agent systems

Evolution of multi-agent teams has been shown to be an effective method of solving complex problems involving the exploration of an unknown problem space. These autonomous and heterogeneous agents are able to go places where humans are unable to go and perform tasks that would be otherwise dangerous or impossible to complete. This research tests the ability of the Orthogonal Evolution of Teams (OET) algorithm to evolve heterogeneous teams of agents which can change their composition, i.e. the numbers of each type of agent on a team. The results showed that OET could effectively produce both the correct team composition and a team for that composition that was competitive with teams evolved with OET where the composition was fixed a priori. Categories and Subject Descriptors I.2.9 [Artificial Intelligence]: Robotics - Autonomous vehicles; I.2.2 [Artificial Intelligence]: Automatic Programming General Terms Algorithms, Design Keywords autonomous vehicles, cooperation, teams

• Type: Conference
• Year: 2009
• Where: GECCO
• Authors: Joshua Rubini, Robert B. Heckendorn, Terence Soule
• Keywords: Artificial Intelligence, Autonomous Vehicles, Correct Team Composition, GECCO 2009, Teams