Improving Indoor Localization Accuracy by Using an Efficient Implicit Neural Map Representation

2025-03-30

AI Summary

Conventional occupancy grid maps limit global localization accuracy for indoor mobile robots operating in known environments. Method: This paper proposes a lightweight implicit neural map representation, the first to jointly model non-projective signed distance fields (SDFs) and orientation-aware projection distances within a LiDAR-based 2D localization framework. Integrated into an enhanced Monte Carlo localization (MCL) pipeline, it enables efficient probabilistic pose estimation. Leveraging LiDAR geometric feature encoding and a compact neural network architecture, the method significantly improves observation model discriminability. Contribution/Results: It achieves real-time performance (>30 Hz), supporting both post-convergence tracking and near-real-time global relocalization. Evaluated on public benchmarks, it outperforms both traditional grid maps and state-of-the-art neural mapping approaches in localization accuracy, thereby overcoming a key bottleneck in deploying implicit neural maps for real-time robot localization.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Learning with Manifolds
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Globally localizing a mobile robot in a known map is often a foundation for enabling robots to navigate and operate autonomously. In indoor environments, traditional Monte Carlo localization based on occupancy grid maps is considered the gold standard, but its accuracy is limited by the representation capabilities of the occupancy grid map. In this paper, we address the problem of building an effective map representation that allows to accurately perform probabilistic global localization. To this end, we propose an implicit neural map representation that is able to capture positional and directional geometric features from 2D LiDAR scans to efficiently represent the environment and learn a neural network that is able to predict both, the non-projective signed distance and a direction-aware projective distance for an arbitrary point in the mapped environment. This combination of neural map representation with a light-weight neural network allows us to design an efficient observation model within a conventional Monte Carlo localization framework for pose estimation of a robot in real time. We evaluated our approach to indoor localization on a publicly available dataset for global localization and the experimental results indicate that our approach is able to more accurately localize a mobile robot than other localization approaches employing occupancy or existing neural map representations. In contrast to other approaches employing an implicit neural map representation for 2D LiDAR localization, our approach allows to perform real-time pose tracking after convergence and near real-time global localization. The code of our approach is available at: https://github.com/PRBonn/enm-mcl.

Problem

• Improving indoor localization accuracy with neural maps
• Enhancing robot pose estimation in real-time environments
• Overcoming limitations of traditional occupancy grid maps

Innovation

• Implicit neural map representation for LiDAR
• Combines signed and projective distance prediction
• Real-time pose tracking in Monte Carlo framework