CURL-SLAM: Continuous and Compact LiDAR Mapping

2025-06-26

AI Summary

To address the high memory overhead and difficulty in balancing compactness with global consistency in traditional LiDAR SLAM, this paper proposes a continuously updatable, compact map representation based on spherical harmonic (SH) implicit encoding. Methodologically, it introduces SH functions to LiDAR SLAM for the first time, enabling a variable-density, continuously differentiable implicit map; further, it designs a joint optimization framework integrating pose estimation and map refinement, employing CURL-based ultra-compact parameterization, non-ICP pose optimization, and local bundle adjustment to achieve 10 Hz real-time performance on CPU. Experiments demonstrate that the method maintains state-of-the-art mapping quality and competitive trajectory accuracy while significantly reducing memory consumption and ensuring global consistency in large-scale environments.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

This paper studies 3D LiDAR mapping with a focus on developing an updatable and localizable map representation that enables continuity, compactness and consistency in 3D maps. Traditional LiDAR Simultaneous Localization and Mapping (SLAM) systems often rely on 3D point cloud maps, which typically require extensive storage to preserve structural details in large-scale environments. In this paper, we propose a novel paradigm for LiDAR SLAM by leveraging the Continuous and Ultra-compact Representation of LiDAR (CURL) introduced in [1]. Our proposed LiDAR mapping approach, CURL-SLAM, produces compact 3D maps capable of continuous reconstruction at variable densities using CURL's spherical harmonics implicit encoding, and achieves global map consistency after loop closure. Unlike popular Iterative Closest Point (ICP)-based LiDAR odometry techniques, CURL-SLAM formulates LiDAR pose estimation as a unique optimization problem tailored for CURL and extends it to local Bundle Adjustment (BA), enabling simultaneous pose refinement and map correction. Experimental results demonstrate that CURL-SLAM achieves state-of-the-art 3D mapping quality and competitive LiDAR trajectory accuracy, delivering sensor-rate real-time performance (10 Hz) on a CPU. We will release the CURL-SLAM implementation to the community.

Problem

• Develops updatable LiDAR maps for continuity and compactness
• Reduces storage needs in large-scale 3D mapping
• Ensures global map consistency after loop closure

Innovation

• Uses spherical harmonics implicit encoding
• Formulates pose estimation as optimization
• Enables real-time performance on CPU