Cycle ProjectsCycle Projects

Dynamic Human-Robot Teams

(Click photo above to visit the project website)

Dynamic Human-Robot Teams

Pickup teams of robots and humans can adapt to unknown, dynamic and adversarial environments, and efficiently execute complex tasks in a robust manner. Both human and robot team members may have only minimal knowledge of each other's behavior but are able to team effectively. Human members might plan while robots might sense and execute tasks. With this system developed in this project, dynamic and unknown obstacles, robot malfunctions, disruptions in communication, and changes in team resources were all handled gracefully and efficiently with team members adopting different roles best suited to efficiently executing the overall mission under the prevailing conditions. We focused specifically on the problem of dynamically forming teams of heterogeneous robots with a range of capabilities together with human team members. The human members of the team were responsible for the highest-level planning and the robots had comparatively limited capabilities: they could sense information about their environment, and they could be tasked to execute abstract tasks. Furthermore, the robots were capable of accepting spoken orders that specified high-level objectives, received important cue information about the terrain, and provided status reports as needed. These research objectives were demonstrated and evaluated in a treasure hunt scenario.

To learn more, visit the project webpage