Adaptive Redirection: A Context-Aware Redirected Walking Meta-Strategy.

Area of Science:

• Virtual Reality
• Human-Computer Interaction
• Computer Science

Background:

• Redirected walking enables natural locomotion in virtual reality (VR) within limited physical spaces.
• Previous research focused on perceptual illusions or individual redirection techniques, lacking a unified system.

Purpose of the Study:

• To formalize a complete redirected walking system by establishing theoretical foundations.
• To introduce adaptive redirection, a meta-strategy combining multiple techniques for enhanced VR exploration.
• To present a novel static planning strategy, COPPER, for optimizing virtual paths.

Main Methods:

• Developed a unified framework for adaptive redirection, integrating various strategies.
• Introduced the Combinatorially Optimized Pre-Planned Exploration Redirector (COPPER) static planning strategy.
• Utilized simulation-based experiments and machine learning to determine empirical adaptation rules.

Main Results:

• Demonstrated that adaptation rules can be empirically determined using machine learning.
• Showcased the effectiveness of partitioning contexts for optimal strategy selection.
• Validated the theoretical framework for adaptive redirection in virtual environments.

Conclusions:

• Adaptive redirection provides a foundational framework for practical implementation of redirected walking.
• The system can be extended to incorporate future techniques for improved performance.
• This approach enhances the exploration of large virtual environments through natural locomotion.