Project Type:

Project

Project Sponsors:

  • US Air Force Office of Scientific Research

Project Award:

  • $1,525,261

Project Timeline:

2020-11-01 – 2024-08-31



Lead Principal Investigator:



Project Team:

Dynamics of Trust Evolution and Calibration: A Field Study of Heterogeneous Human-Machine Teams with High Levels of Autonomy Operating in Contexts with Real Users, Real Systems, and Real Consequences


Project Type:

Project

Project Sponsors:

  • US Air Force Office of Scientific Research

Project Award:

  • $1,525,261

Project Timeline:

2020-11-01 – 2024-08-31


Lead Principal Investigator:



Project Team:

In the past decade, research on human machine team (HMT) trust has burgeoned. However, most studies have been limited to laboratory/simulation-based environments with a single operator interacting with the machine. While such studies are valuable in exploring solution approaches, there is a lack of research studying trust evolution and calibration in real systems (R3 ? real users, real system, real consequence), that is, in a heterogeneous team of multiple real highly-autonomous machine teammates and real human teammates operating in situations with real consequences (e.g., life or death, or loss of a critical mission and/or valuable assets). Examples of the heterogeneity in these HMT teams include different collaboration duration (e.g., permanent or temporary), collocation (i.e., some members are collocated while others are non-collocated), vehicle types (e.g., aerial or ground), and capabilities (e.g., different autonomy levels or variable communication links and delays). We propose to conduct a 5-year field study to: 1) Obtain foundational lessons and insights on how trust evolves and is calibrated from the time the heterogeneous HMT is deployed until the trust level reaches a steady value (if at all); 2) Identify how non-technology and technology factors influence the trust evolution and then define its essential elements (i.e., rise/settling time, nature of fluctuation, steady states); 3) Utilize the information and insights gained to validate extant theoretical trust models (e.g., Lee & See model) against trust evolution of R3 heterogeneous HMTs and extend/adapt the models; and 4)Formulate research questions and recommend pathways to translate the results of this proposed basic research project to applied research. We will use specific heterogeneous HMTs as the context for a field study based on R3 considerations. The research will focus on three extant, cutting edge HMTs with autonomy technologies that have great potential relevance to Air Force missions and to which our research team has access, including the Jet Propulsion Laboratory?s (JPL) Mars Helicopter, JPL?s DARPA SubTerranean (SubT), and NASA Armstrong Flight Research Center?s (AFRC) Traveler with Expandable Variable-Autonomy Architecture. The time period suggested is a 5-year field study utilizing Rapid Assessment Procedures (RAP) Method. In project year 1-5, we will employ semi-structured interviews and surveys to facilitate the use of grounded theory1 as an inductive, hypothesis-generating method and to provide data for deductive, qualitative analysis. We anticipate involving 30-40 participants organized by various three HMT teams. In project years 3-5, we will use the lessons learned and insights gained to validate and extend or adapt extant theoretical trust models. Additionally, in project years 2-5, we will work closely with the AFOSR Program manager to formulate precise questions that will guide the application basic research results of the proposed project to future applied research efforts. The project?s potential contributions to the Air Force?s missions include: 1) guidelines/recommendations for designers of future Air Force systems that involve heterogeneous HMTs with high levels of autonomy; 2) transition research results to applied research for AFRL Man-Unmanned Teaming (MuM-T) Concepts of Operations (ConOps); and 3) contribute to ConOps that involve heterogeneous HMTs and related advanced autonomy technologies (e.g., drone control algorithms, system health management, autonomy algorithms for experimental satellite systems).






Give Feedback