Exploring Perception-Based Techniques for Redirected Walking in VR: A Comprehensive Survey

TL;DR

This survey tackles the problem of enabling real walking in VR when physical space is limited by introducing a perception-based RDW taxonomy centered on Target Orientation Calculation. It analyzes 165 high-impact papers across gains, gain application, orientation strategies, and enhancements to construct a modular framework that clarifies how redirection is computed, applied, and aligned with the physical and virtual spaces. Key contributions include a novel Target Orientation Calculation component, an extensive component-wise framework for constructing RDW techniques, and insights into underexplored areas such as predictive-alignment and multi-user fairness, backed by data-driven observations. The findings highlight that steering and reactive gains dominate the current literature, while alignment and predictive methods remain underexplored, suggesting new directions for designing RDW systems that improve presence, spatial knowledge, and user experience in varied VR contexts.

Abstract

We present a comprehensive survey of perception-based redirected walking (RDW) techniques in virtual reality (VR), presenting a taxonomy that serves as a framework for understanding and designing RDW algorithms. RDW enables users to explore virtual environments (VEs) larger than their physical space, addressing the constraints of real walking in limited home VR setups. Our review spans 232 papers, with 165 included in the final analysis. We categorize perception-based RDW techniques based on gains, gain application, target orientation calculation, and optional general enhancements, identifying key patterns and relationships. We present data on how current work aligns within this classification system and suggest how this data can guide future work into areas that are relatively under explored. This taxonomy clarifies perception-based RDW techniques, guiding the design and application of RDW systems, and suggests future research directions to enhance VR user experience.

Figures (4)

• Figure 1: Taxonomy of Redirected Walking Algorithms – RDW algorithms consist of three core components: (1) locomotion gains, (2) methods for applying those gains, and (3) orientation calculation. Gains vary by detectability (overt/subtle) and type (rotation, translation, curvature, bending), with types combinable. Gain application methods are categorized as reactive, predictive, or scripted. Orientation strategies include steering, avoidance, and alignment. Optional enhancements address multi-user scenarios, complex spaces, gain masking, motion diversity, and immersion.
• Figure 2: Steering and Avoidance Target Orientation Calculation Method Examples: a) steering to a single target. b) steering to a path. c) steering to multiple targets. d) avoidance with a path avoiding multiple obstacles.
• Figure 3: Alignment Target Orientation Calculation Method Example: In this method the alignment is defined by the overlap of open space in view, as such the target orientation is determined by finding what orientation will increase the overlap from the current state.
• Figure 4: (a) Percentage of works on RDW methods by combination of Gain Application type and Target Orientation Calculation method. (b) Percentage of works by Target Orientation Calculation type. (c) Percentage of works by Gain Application type