UMLoc: Uncertainty-Aware Map-Constrained Inertial Localization with Quantified Bounds

2026-01-10

arXiv.org

AI Summary

This work addresses the challenge of significant drift accumulation in pure inertial navigation under GPS-denied environments due to sensor noise and bias. To mitigate this, the authors propose UMLoc, a novel end-to-end localization framework that, for the first time, integrates IMU uncertainty quantification with map constraints. Specifically, the system employs LSTM-based quantile regression to estimate well-calibrated positional confidence intervals and introduces a conditional generative adversarial network augmented with cross-attention mechanisms to fuse IMU temporal dynamics with distance-based floorplan priors, thereby generating geometrically consistent and drift-resilient trajectories. Experimental results demonstrate that UMLoc achieves an average drift ratio of 5.9% over 70-meter trajectories, with an absolute trajectory error (ATE) of 1.36 meters, while also exhibiting well-calibrated predictive uncertainty across three benchmark datasets.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Calibration & Uncertainty Quantification
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Inertial localization is particularly valuable in GPS-denied environments such as indoors. However, localization using only Inertial Measurement Units (IMUs) suffers from drift caused by motion-process noise and sensor biases. This paper introduces Uncertainty-aware Map-constrained Inertial Localization (UMLoc), an end-to-end framework that jointly models IMU uncertainty and map constraints to achieve drift-resilient positioning. UMLoc integrates two coupled modules: (1) a Long Short-Term Memory (LSTM) quantile regressor, which estimates the specific quantiles needed to define 68%, 90%, and 95% prediction intervals serving as a measure of localization uncertainty and (2) a Conditioned Generative Adversarial Network (CGAN) with cross-attention that fuses IMU dynamic data with distance-based floor-plan maps to generate geometrically feasible trajectories. The modules are trained jointly, allowing uncertainty estimates to propagate through the CGAN during trajectory generation. UMLoc was evaluated on three datasets, including a newly collected 2-hour indoor benchmark with time-aligned IMU data, ground-truth poses and floor-plan maps. Results show that the method achieves a mean drift ratio of 5.9% over a 70 m travel distance and an average Absolute Trajectory Error (ATE) of 1.36 m, while maintaining calibrated prediction bounds.

Problem

• inertial localization
• GPS-denied environments
• drift
• IMU
• localization uncertainty

Innovation

• Uncertainty-aware localization
• Map-constrained IMU
• Quantile regression
• Conditional GAN
• Drift-resilient inertial navigation