Understanding, Modeling and Simulating Unintended Positional Drift during Repetitive Steering Navigation Tasks in...

Area of Science:

• Human-Computer Interaction
• Virtual Reality
• Usability Engineering

Background:

• Virtual steering techniques allow navigation in large virtual environments (VEs) beyond physical workspace limits.
• Users may exhibit Unintended Positional Drift (UPD) during virtual steering, posing safety risks by approaching physical boundaries.

Purpose of the Study:

• To analyze and model Unintended Positional Drift (UPD) during virtual navigation tasks.
• To provide a foundational understanding for designing safer virtual steering techniques.

Main Methods:

• Collected positional and orientation data from 18 users performing a virtual slalom task with torso-steering.
• Developed two UPD models: linear regression and Gaussian Mixture Model (GMM).
• Evaluated models using simulation-based assessments with virtual agents.

Main Results:

• Characterized UPD patterns during virtual steering across different trajectory curvatures.
• Demonstrated the feasibility of simulation-based evaluations for studying UPD.
• Identified distinct UPD behaviors influenced by virtual environment navigation.

Conclusions:

• The study successfully models Unintended Positional Drift (UPD) in virtual steering.
• Simulation-based evaluations are effective for analyzing UPD.
• Findings can inform the design of virtual reality navigation systems to mitigate UPD and enhance user safety.