Presentation
Analyzing Human-Automation Interdependence Based on the Teammate, Situational Properties, and Interaction Sequence
DescriptionAutomation’s imperfection requires teaming with drivers for certain tasks, creating interdependent relationships where both parties influence each other. Therefore, it is important to support driver-automation interdependence to align their behaviors with team goals. We expand the Interdependence Analysis (IA) method to evaluate interdependent relationships within driver-automation teams, considering their situational contexts (e.g., cost of task failure) and the timescales of team tasks. By analyzing driver-automation teaming in a lane-changing scenario, our proposed IA offered suggestions to improve the collaboration. Specifically, adding collaboration features can enhance the trajectory planning aspect of lane-changing, which carries a high cost of failure. Timescale decomposition further suggested that trajectory planning relies on short-term road user movement and lane predictions—tasks solely assigned to automation but where the driver can offer support. This study demonstrates the benefits of considering situational contexts and task timescales in analyzing interdependence within driver-automation teams and designing supportive strategies for them.
Contributors
Postdoctoral Researcher
Research Scientist
Event Type
Industry/Practitioner Case Study
Lecture
TimeFriday, September 13th10:05am - 10:25am MST
LocationFLW Salon G
Surface Transportation