In this paper, we consider the problem of environmental constraints on teams and examine how the constraints impact teamwork. We simulate a realistic problem for teams in the physical world by implementing a limit on the range of communication among agents. Using BlocksWorld for Teams from Delft University of Technology, we build teams of 3-4 agents with varying communication ranges and of different agent type (regular task-performer, communication-only or reconnaissance-only) to test team performance in the constrained environment. We analyze the results and discuss implications to the team mental model. We show that adding a helper agent to the team can reduce task completion time, but not for all types of agents and environments. © 2013 Springer-Verlag.
CITATION STYLE
Manner, M. D., & Gini, M. (2013). Improving agent team performance through helper agents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7764 LNAI, pp. 89–105). https://doi.org/10.1007/978-3-642-36444-0_6
Mendeley helps you to discover research relevant for your work.