V2I Work Zone Geometry Reconstruction with Pose-Conditioned UWB Range Denoising

2026-05-27

AI Summary

This work addresses the unreliability of geometric reconstruction in construction zones caused by sporadic outliers, non-line-of-sight (NLOS) errors, disordered anchor sequences, and vehicle pose uncertainty in ultra-wideband (UWB) ranging. To tackle these challenges, the authors propose a pose-conditioned, permutation-equivariant denoising model that integrates vehicle pose as a geometric prior. The method leverages anchor-level temporal prediction, symmetric set aggregation, and pose-conditioned residual decoding to effectively handle unordered and missing anchors. A two-stage training strategy with NLOS-aware weighted supervision is employed to enhance robustness. Evaluated in both real-world vehicle-to-infrastructure (V2I) scenarios and simulations, the approach reduces reconstruction error by 66.9%, significantly improving ranging accuracy and cone localization performance while demonstrating strong resilience to anchor reordering and partial anchor loss.

Abstract

Reliable work zone mapping is important for connected and autonomous vehicles (CAVs) to navigate safely and smoothly through work zone areas. Cone-mounted ultra-wideband (UWB) roadside units (RSU) offer a cost-effective way for work zone layout inference, as roadside anchors and vehicle tags provide direct vehicle-to-infrastructure (V2I) range constraints for work zone geometry reconstruction. However, UWB range estimation is degraded by bursty outliers, non-line-of-sight (NLOS) errors, arbitrary anchor-ordering issues, and vehicle pose uncertainties in practical field deployments. To address these challenges, this study proposes a pose-conditioned, permutation-equivariant predictive denoiser for multi-anchor UWB ranging. The model employs shared anchor-wise temporal prediction to capture range dynamics, symmetric set aggregation to handle unordered and missing anchors, and pose-conditioned residual decoding to incorporate vehicle motion as a geometric prior. A two-stage training strategy first learns prediction from observed ranges, and then fine-tunes the denoiser with NLOS-weighted supervision. The method is evaluated on rare real-world V2I UWB field data collected with a CAV, as well as on controlled large-scale simulation benchmarks for ablative insights. Results show that the proposed method substantially improves range accuracy, cone localization, and work zone geometry reconstruction in challenging NLOS-dominated regimes, remains robust to anchor re-indexing and moderate anchor dropout, and reduces measurement-weighted field MSE by 66.9% relative to the raw input.

Problem

• UWB ranging
• work zone geometry reconstruction
• non-line-of-sight (NLOS)
• vehicle-to-infrastructure (V2I)
• pose uncertainty

Innovation

• pose-conditioned denoising
• permutation-equivariant
• UWB ranging
• V2I
• work zone reconstruction