CS Paper Radar

Endless VR Walk: Dynamic Mapping of Limited Physical Spaces into Infinite Linear Paths

Sofia Martinez, Reagan Haddix, Miles Haase, Angelos Barmpoutis

cs.HC ·submitted 2026-11-01 03:27 UTC · v1
published

Abstract

Natural locomotion in virtual reality (VR) can provide a more immersive and engaging experience, but the physical space required for unrestricted walking is a significant limitation. This paper presents Endless VR Walk, a system for dynamically mapping a limited physical space into an effectively infinite linear virtual path. The approach uses a recursive path-generation algorithm to construct virtual blocks in real time as the user walks, while constraining the path to the available physical area and preventing spatial overlap. We evaluate path variability under different physical-space constraints and identify the smallest space that enables diverse path configurations. The resulting framework provides a platform for applications benefiting from extended walking, including physical therapy, rehabilitation, and fitness, as well as adaptations of traditional platformer and endless-runner games in VR.

Comment
2026 ACM SIGGRAPH Conference on Motion, Interaction and Games

Closest papers

Nearest neighbours by meaning across the whole index.

TRAILS: Trajectory Generation for Automated Indoor Locomotion Studies

cs.RO ·2026-11-02 ·Juho Kalliokoski, Evan G. Center, Steven M. LaValle, Timo Ojala, Basak Sakcak
code/data similarity 0.64
Abstract · 8 pages, 4 figures, IEEE Telepresence 2026

Virtual reality experiments involving navigation and simulated robot motion often require large, realistic environments and numerous distinct trajectories to avoid repetitive navigation while providing sufficient variation for controlled experiments. Designing such scenarios manually is labor-intensive, particularly when trajectories must satisfy predefined geometric and kinematic constraints while remaining diverse and comparable across experimental conditions. This paper presents TRAILS, an automated framework for generating complete virtual navigation scenarios consisting of an environment layout, a path, and a time-parameterized motion profile. The framework combines a constraint-based procedural environment generator inspired by Wave Function Collapse, trajectory generation using optimized cubic Bézier curves, and speed-profile generation subject to kinematic constraints. The generated scenarios provide diverse indoor environments with corridors, open spaces, and obstacles, together with trajectories of identical duration that vary in path geometry, wall proximity, and velocity profiles while minimizing systematic dependencies between these characteristics. The framework is designed as an easy-to-use tool requiring only a small number of intuitive user-defined parameters to automatically generate large sets of distinct navigation scenarios satisfying prescribed geometric and kinematic constraints. By relying on explicit, seed-controlled algorithmic procedures rather than data-driven models, the generated scenarios remain transparent, reproducible, and configurable, making the framework well suited for telepresence and human-subject experiments. The code is available at https://github.com/JucciK/Trails .

0.34

Sensing Instability, Adapting the Scene: A Real-Time Movement-Smoothing Design Framework for Stable VR Locomotion

cs.HC ·2026-09-30 ·Ramisa Fariha Joyee, M. Rasel Mahmud
published similarity 0.62
Abstract · 3 pages, accepted as ISMAR 2026 AXR workshop paper

Users experience different balance challenges while standing, walking, and turning in virtual reality (VR), yet most locomotion techniques apply the same visual behavior regardless of movement state. We present a real-time movement-smoothing design framework in this paper that organizes visual adaptations according to movement-specific balance demands. The framework introduces three design strategies targeting standing, walking, and turning, illustrating how state-aware visual adaptations can support postural stability during locomotion. This paper focuses on the framework design and implementation, while a user study is planned as future work. Our work guides the development of future context-aware VR locomotion systems that better support safe and comfortable navigation.

0.38

Preliminary Evaluation of Transition-Aware Controller Locomotion Adaptations for Supporting Postural Stability in VR

cs.HC ·2026-09-30 ·Ramisa Fariha Joyee, M. Rasel Mahmud
published similarity 0.55
Abstract · 4 pages, 3 figures, accepted in ISMAR 2026 poster track

We present three locomotion adaptation approaches: Motion Acceleration, Turn Acceleration, and Motion Deceleration to improve postural stability during body-state transitions in virtual reality (VR). The system detects standing-to-walking, turning, and walking-to-stopping transitions and applies adaptive locomotion smoothing. Motion Acceleration gradually increases locomotion speed when users begin walking, Turn Acceleration smooths rotation while turning, and Motion Deceleration gradually reduces movement speed before stopping. We evaluated these techniques in a virtual navigation task using objective and subjective balance measures. Preliminary results show reduced center of pressure (COP) velocity and improved balance confidence. These findings suggest that locomotion adaptations can improve balance and navigation experience.

0.44

Make it Walk: Computational Design of LEGO Walkers

cs.RO ·2026-10-21 ·Siqi Li, Fancheng Gao, Chi-Wing Fu, Peng Song
similarity 0.54
Abstract

LEGO walkers are walking automata constructed from modular LEGO Technic bricks. They exhibit intriguing locomotion and are also highly reconfigurable through the modular Technic bricks, making them widely applicable in education, recreation, and robotics research. However, existing LEGO walkers are designed mostly by reusing existing single-actuator walking mechanisms. Although the locomotion is stable, their fixed walking gaits limit their adaptability and versatility. This paper presents the first computational solution capable of designing LEGO walkers that not only walk stably, but also support a variety of walking gaits. Our idea is twofold. First, we control the foot motion trajectory of each leg mechanically via a specifically-designed gear-linkage mechanism, in which we incorporate a pair of 3D-printed non-circular gears to work with a LEGO Technic mechanism. Second, we drive each leg by an independent motor and synchronize the timings of multiple leg motions via a simple yet adaptable control algorithm, enabling stable and versatile locomotion. Importantly, we take this modeling as a foundation to co-design the morphology and control of the LEGO walkers to meet the target locomotion and appearance. Leveraging our approach, we created a variety of LEGO walkers, demonstrating various walking gaits and appearances, and fabricated some of them to show their locomotion capability in the real world.

0.22

Geometry-Aware Predictive Control for Walk-to-Run Bipedal Obstacle Crossing

cs.RO ·2026-10-11 ·Aashish Sahu, R. Prasanth Kumar
similarity 0.54
Abstract

This paper presents a geometry-aware predictive control framework for bipedal obstacle crossing across walking and running regimes. The proposed approach integrates a phase-structured trajectory generator with a constrained model predictive control (MPC) tracking layer to generate coordinated pelvis and swing-foot motions during obstacle traversal. The crossing process is represented through five interpretable phases: pre-crossing, leading-leg clearance, straddle, trailing-leg clearance, and recovery. Within this structure, swing-foot elevation is adapted as a function of obstacle height, while lateral stance widening and pelvis repositioning are introduced as functions of obstacle depth to improve transverse clearance and body alignment. A continuous gait-parameter evolution enables smooth walk-to-run transition without discrete controller switching. The generated references are tracked using MPC to regulate motion smoothness, clearance feasibility, touchdown behavior, and lateral admissibility. The framework is evaluated in reduced-order simulation over five progressively difficult obstacles under walking and running conditions. The robot successfully crosses the first four obstacles, achieving minimum body-obstacle clearances of 0.036, 0.036, 0.059, and 0.017 m, respectively. The fifth and most demanding obstacle leads to collision, identifying the current operating boundary of the proposed reduced-order formulation. Overall, the method provides a compact, interpretable, and computationally efficient basis for intelligent obstacle-aware bipedal locomotion.

0.24

Learning Scene-Aware Humanoid Locomotion through 3D Clutter from Immersive Human Demonstrations

cs.RO ·2026-09-17 ·Beichen Wang, Tong Xu, Daniel Kosukhin, Yuen-Hei Yeung, Yuanjie Lu +1 more
similarity 0.53
Abstract

While learning from human motions has enabled highly dynamic humanoid skills such as dancing and martial arts in obstacle-free space, traversal through densely cluttered environments remains underexplored. These spaces are three-dimensional and geometrically constrained, requiring scene-aware locomotion that tightly couples whole-body motion with scene geometry for obstacle avoidance. To address these challenges, we present Moving Through Clutter (MTC), a learning-from-demonstration framework for scene-aware humanoid locomotion. To bypass costly physical scene construction, MTC uses procedurally generated Virtual Reality environments for immersive data collection. To transform these human motions into training-ready humanoid motions, we propose a scene-aware motion retargeting algorithm that converts human demonstrations into humanoid trajectories while strictly enforcing robot-scene clearance to guarantee collision-free traversal. These reference trajectories are then used to train a scene-aware locomotion policy that deploys on a Unitree G1 humanoid. Evaluated on our proposed MTC-Challenge for multi-obstacle traversal, the policy demonstrates a 70.2% collision-free rate across diverse scenarios, successfully traversing complex environments through diverse whole-body skills, including crawling through low-clearance passages and squeezing through narrow gaps.

0.21

REACH: An RL-based Hierarchical Humanoid Crawling Framework for Adaptable Confined Space Traversal

cs.RO cs.AI ·2026-10-27 ·Shengzhang Xu, Yuning Guo, Weiyi Hu, Yifeng Xu, Zhongyu Li +1 more
publishedcode/data similarity 0.53
Abstract · To appear at IEEE Humanoids 2026; Project page https://aislab-sustech.github.io/REACH/

Humanoid locomotion in confined spaces requires continuous height adaptation across multiple locomotion modes, yet most existing approaches remain limited to height-adjustable bipedal walking. We present REACH, an RL-based hierarchical framework that enables a Unitree G1 humanoid to traverse height-constrained environments through variable-height bipedal walking and crawling. The learned policies regulate body height over complementary command ranges of 1.27--0.50m in bipedal locomotion and 0.50--0.20m in crawling, extending traversal beyond the clearance limits of crouched walking. To address contact-rich crawling dynamics, we employ a history-conditioned latent representation that captures terrain and interaction context from proprioceptive observations. A bidirectional transition policy further connects crouched bipedal locomotion and crawling across large kinematic and contact discontinuities. These skills are integrated with a LiDAR-based planner that converts point clouds into collision-free three-dimensional paths and locomotion commands, including proactive height targets. Real-world experiments demonstrate accurate height and velocity tracking, robust terrain traversal, reliable mode transitions and confined-space traversal with the proposed pipeline. Code and videos are available at https://aislab-sustech.github.io/REACH/.

0.30

SHOES: Sim-to-real Hardware-adaptation via On-policy Expansion of Skills for Legged Robots

cs.RO ·2026-10-12 ·Kota Tsuruhara, Tsung-Yuan Chen, Kento Kawaharazuka
published similarity 0.52
Abstract · Accepted at the 10th Conference on Robot Learning (CoRL 2026)

Humans expand their locomotion capability by wearing appropriate footwear, such as hiking boots for mountains or skates for ice. Footwear changes the ground-contact interface and enables traversal of otherwise difficult terrain. In contrast, locomotion controllers are typically trained for a fixed morphology, and their performance can degrade when shoes alter foot geometry, mass, and friction. We propose SHOES, a sample-efficient framework for adapting locomotion policies to footwear-induced morphology changes. SHOES freezes a simulation-trained nominal policy and fine-tunes only lightweight, identity-initialized modulation modules on the robot, preserving the initial behavior while enabling efficient hardware adaptation. We also introduce SafeBox, a workspace-aware waypoint sampler that constrains online adaptation within a user-defined workspace. The simulation results with a shoe-equipped Unitree Go2 show that SHOES substantially recovers locomotion performance after footwear-induced morphology changes using limited additional adaptation. Under the same adaptation budget, SHOES outperforms FiLM-only, Linear Adapter-only, and residual-RL baselines. Real-world experiments on a Unitree Go2 fitted with human shoes further evaluate the effectiveness of SHOES for hardware deployment.

0.21