A Mobile Data Collection System for Studying Human Autonomy Teaming in Conjunction with Passive Context and Psychophysiological Sensing
Abstract
It is expected that the ability for autonomous agents to adapt to their human teammates will be a critical development in building effective human-autonomy teams for the future. We argue that the current state of research in this area is limited by high experimenter and participant burden (e.g., traveling to a laboratory location; single, multi-hour experimental sessions; expensive, custom experimental set-ups). As an alternative, we develop a data collection system that makes use of a mobile videogame platform in conjunction with passive sensing of context and psychophysiology through the phone and a wearable device. This design dramatically reduces burden while maintaining flexible experimental design and multi-modal measurement. Here, we describe this novel system, as well as an illustrative research design for how it would be used to develop autonomous agents with adaptive capabilities. We also present a small set of pilot data validating several aspects of the system.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2021
- Accession Number
- AD1156085
Entities
People
- Andrew Campbell
- Evan Carter
- Lydia Tapia
- Torin Adamson
- Weichen Wang
- Yazied Hasan
Organizations
- Dartmouth College
- United States Army
- University of New Mexico