Embedded Files
The MULTI-UAV Simulation for Teaming Research and Evaluation of Autonomous Missions (STREAM) is used to explore how humans respond to increasing agent decision-making capabilities in multi-agent teaming environments.
The MULTI-UAV Simulation for Teaming Research and Evaluation of Autonomous Missions (STREAM) is used to explore how humans respond to increasing agent decision-making capabilities in multi-agent teaming environments.
Images of the simulation (top), program interface (bottom left), and street views (bottom right).
Summary
This project was created with researchers from Florida Tech's Atlas Lab under the Consortium Research Fellows Program (CRFP).
The simulation allows researchers to assess the impact of agent autonomy on human performance, behavior, trust, stress, and workload.
Features and Design
Software
Unreal Engine 4
_
Contributions
Team:
Tyler Frost
Jerry Huggins
_
Collaborators:
Atlas Lab, Florida Tech
Summer Rebensky, PhD
Kendall Carmody
Cherrise Ficke
Outcomes and Additional Information
The experiment asked participants to complete a series of four intelligence surveillance and reconnaissance missions with agents of four levels of autonomy. Participants worked with four agent teammates to take pictures of soldiers as targets and mark them as friendly, enemies, or not targets.
When a picture was taken (either by the participant or the agent), the participant would need to decide on what type of target it was. The participant received varying levels of assistance with this decision-making task based on the level of autonomy condition. After each mission, participants filled out a series of surveys while performance information was collected automatically by the simulator.
Participant’s performance, stress, and workload scores indicated that higher levels of autonomy resulted in lower levels of stress and workload, and thus better performance, but can lead to automation complacency.
Multi-UAV S.T.R.E.A.M Testbed.pdfPublications related to this effort
Rebensky, S., Carmody, K., Ficke, C., Carroll, M., & Bennett, W. (2022). Teammates instead of tools: The impacts of level of autonomy on mission performance and human–agent teaming dynamics in multi-agent distributed teams. Frontiers in Robotics and AI, 9.
Carmody, K., Ficke, C., Nguyen, D., Addis, A., Rebensky, S., & Carroll, M. (2022). A qualitative analysis of trust dynamics in human-agent teams (HATS). Proceedings of The Human Factors and Ergonomics Society (HFES), Atlanta, GA.
Ficke, C. (2022). The examination of factors that influence trust in a multi-agent team context [Thesis]. Florida Institute of Technology.
Rebensky, S., Carmody, K., Ficke, C., Nguyen, D., Carroll, M., Wildman, J., & Thayer, A. (2021). Whoops! Something went wrong: Errors, trust, and trust repair strategies in human agent teaming. Human Computer Interaction International 2021.
Google Sites
Report abuse