Dynamically Formed Heterogeneous Robot Teams Performing TightlyCoordinated Tasks.pdfVIP

Dynamically Formed Heterogeneous Robot Teams Performing Tightly-Coordinated Tasks E. Gil Jones, Brett Browning, M. Bernardine Dias, Brenna Argall, Manuela Veloso Anthony Stentz School of Computer Science Carnegie Mellon University Pittsburgh, Pennsylvania 15213 Email: {egjones,brettb,mbdias,bargall,mmv,axs}@cs.cmu.edu Abstract— As we progress towards a world where robots play community has addressed the coordination of simulated agents an integral role in society, a critical problem that remains to be built by different groups [5], none has chosen to address solved is the Pickup Team Challenge; that is, dynamically formed this pickup challenge for the coordination of multiple robots. heterogeneous robot teams executing coordinated tasks where little information is known a priori about the tasks, the robots, We believe this research direction of forming dynamic teams and the environments in which they will operate. Successful will greatly advance the science of multi-robot systems. Thus, solutions to forming pickup teams will enable researchers to the Pickup Team Challenge is to dynamically form teams experiment with larger numbers of robots and enable industry of robots (and eventually humans) given very little a priori to efficiently and cost-effectively integrate new robot technology information. That is, team members may have only minimal with existing legacy teams. In this paper, we define the challenge of pickup teams and propose the treasure hunt domain for prior kno