A Shared Meta-Model Framework to Enable Multi-Directional Reliance for Effective Collaborative Human-Autonomy Teaming

Project begins 2024.

Major challenges in human-autonomy teaming include overwhelmed cognitive loads, a lack of shared understanding, and poor real-time communication among agents (human and autonomous), especially in multi-agent systems (one-to-one, one-to-many, many-to-one, many-to-many). When communications are lost or degraded to a state that is unhelpful or distracting, agents of a team can become overloaded and lose trust in their fellow teammates.

The main objective of the proposed research includes studying the impact a shared mental model-focused framework has on collaborative interactions, team situation awareness, multi-directional reliance, (human) cognitive loads, and team performance through improved shared understanding, real-time communication, and information handling. This impact will be investigated by completing the objectives (refer to timeline below for additional details) and answering the following research questions:

RQ1: How do collaborative interactions and information-sharing influence human/autonomy mental models, shared understanding, team situation awareness, (human) cognitive loads, bi-directional reliance, and task performance?

RQ2: How can teammates interpret (human/autonomy) intent and behavior from shared mental models?

RQ3: How do shared mental models and bi-directional reliance assist with understanding each other’s capabilities, limitations, and preferences?

RQ4: What impact does multi-directional reliance have on a team’s shared understanding, shared situation awareness, collaborative interactions, and task performance?

Publications from Prior Work closely related to the proposed project:

• A. L. Aldridge and C. L. Bethel, “M-OAT Shared Meta-Model Framework for Effective Collaborative Human-Autonomy Teaming,” in Companion of the 2023 ACM/IEEE International Conference on Human-Robot Interaction (HRI ’23), New York, NY, USA: Association of Computing Machinery, 2023, pp. 663–666. Available: https://doi.org/10.1145/3568294.3580169
• B. Mooers, A. L. Aldridge, A. Buck, C. L. Bethel, and D. T. Anderson, “Human-robot teaming for a cooperative game in a shared partially observable space,” in Proceedings of SPIE Defense + Commercial Sensing 12525, Geospatial Informatics XIII, 125250B, Jun. 2023. Available: https://doi.org/10.1117/12.2663430

#2.20