Location

FIT

Document Type

Event

Keywords

human-automation teamwork, trust in automation

Description

Advancements in robotics, artificial intelligence, and other automation have highlighted the need for humans to work together with machines in a more flexible and collaborative fashion than previously possible. To formulate effective human-robot teams, it is critical to understand which factors play important roles in enhancing human-robot teamwork. To gain preliminary insights into key factors of effective human-robot teams, we carried out an experiment using an enhanced version of the “Lunar Lander” game, where the goal is to safely land a spacecraft on the moon’s surface in concert with an AI teammate. We specifically attempted to observe some patterns of communication across high-performing teams in the experiment. Due to the limited number of experimental participants, the results of the experiment did not definitively identify the factors that account for effective teams. Instead, the experiment indicated potential avenues to further investigate, including intent-oriented communication and trust in teams with human and non-human entities. This paper presents findings from the experiment and discusses future work to extend the scope of the study to include teleoperation of unmanned vehicles with communication delay.

DOI

https://doi.org/10.5038/YEXL3996

Share

COinS
 

Preliminary Insights into Enhancing Human-Robot Teamwork

FIT

Advancements in robotics, artificial intelligence, and other automation have highlighted the need for humans to work together with machines in a more flexible and collaborative fashion than previously possible. To formulate effective human-robot teams, it is critical to understand which factors play important roles in enhancing human-robot teamwork. To gain preliminary insights into key factors of effective human-robot teams, we carried out an experiment using an enhanced version of the “Lunar Lander” game, where the goal is to safely land a spacecraft on the moon’s surface in concert with an AI teammate. We specifically attempted to observe some patterns of communication across high-performing teams in the experiment. Due to the limited number of experimental participants, the results of the experiment did not definitively identify the factors that account for effective teams. Instead, the experiment indicated potential avenues to further investigate, including intent-oriented communication and trust in teams with human and non-human entities. This paper presents findings from the experiment and discusses future work to extend the scope of the study to include teleoperation of unmanned vehicles with communication delay.