GlassGuard: Verified Glass Plane Mapping for Robot Navigation

2026-10-01

AI Summary

This study addresses the challenge that LiDAR beams penetrate glass surfaces, resulting in navigation maps that lack collision boundaries while erroneous reconstructions contaminate traversable space. To overcome this, we propose a glass plane mapping framework leveraging complementary evidence from vision and LiDAR. Specifically, foundation vision models generate glass masks, which are combined with structural 3D cues to formulate planar hypotheses. A depth-free 2D projective geometric verification mechanism is further introduced to balance glass coverage against free-space integrity. We establish a dual evaluation criterion demonstrating that our approach achieves 85% panoramic glass coverage in real-world scenarios while reducing false-positive voxels by 5 to 17 times compared to baseline methods. These results indicate significant improvements over existing techniques, effectively ensuring safer robot navigation in environments containing transparent obstacles.

Abstract

Transparent and specular surfaces pose a serious challenge to LiDAR-based SLAM and navigation because laser returns may pass through glass, leaving collision boundaries absent from the map. Prior work attempts to reconstruct the missing surfaces, but inaccurate obstacle placement can create the opposite failure: contamination of traversable free space. Recognizing this dual requirement, we present GlassGuard, a navigation-oriented framework for reconstructing planar architectural glass from complementary visual and LiDAR evidence. We formulate success in terms of both glass coverage and free-space contamination and apply this principle throughout proposal verification and global map construction. A foundation vision model provides glass-instance masks, structural 3D cues generate metric plane hypotheses, and depth-free 2D projective geometry checks their orientations before they enter a consolidated global map. We evaluate GlassGuard in nine building-scale scenes spanning diverse glass structures, spatial scales, and lighting conditions, with more than one hour and 2.1 km of real-world robot traversal. GlassGuard achieves 85% of total glass coverage for its panoramic version. Under identical pinhole inputs, GlassGuard achieves 82% total coverage, compared with at most 61% for the evaluated baselines, while producing 5-17x fewer false voxels per frame. Qualitative examples with a navigation planner illustrate the reconstructed planes blocking paths through glass while leaving traversable routes open. The project page is available at https://glassguardproject.github.io/.

Problem

• glass detection
• robot navigation
• LiDAR SLAM
• transparent surfaces
• free-space contamination

Innovation

• Glass plane mapping
• Foundation vision model
• SLAM navigation
• Sensor fusion
• Projective geometry