XGrid-Mapping: Explicit Implicit Hybrid Grid Submaps for Efficient Incremental Neural LiDAR Mapping

2025-12-24

AI Summary

To address the weak geometric awareness of implicit methods and poor real-time performance of explicit voxel-based approaches in large-scale incremental neural LiDAR mapping, this paper proposes a hybrid explicit-implicit grid submap framework. Our method introduces: (1) the first explicit sparse grid built upon VDB’s hierarchical sparse structure and submap organization, tightly coupled with implicit neural fields to jointly ensure geometric fidelity and representational capacity; and (2) a knowledge distillation–driven alignment strategy for overlapping regions across submaps, integrated with a dynamic point cloud pruning mechanism to enhance both consistency and computational efficiency. The framework supports long-term incremental deployment. Extensive experiments on multiple benchmarks demonstrate significant improvements over state-of-the-art methods—achieving higher mapping accuracy, faster inference speed, and greater robustness.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• 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

Abstract

Large-scale incremental mapping is fundamental to the development of robust and reliable autonomous systems, as it underpins incremental environmental understanding with sequential inputs for navigation and decision-making. LiDAR is widely used for this purpose due to its accuracy and robustness. Recently, neural LiDAR mapping has shown impressive performance; however, most approaches rely on dense implicit representations and underutilize geometric structure, while existing voxel-guided methods struggle to achieve real-time performance. To address these challenges, we propose XGrid-Mapping, a hybrid grid framework that jointly exploits explicit and implicit representations for efficient neural LiDAR mapping. Specifically, the strategy combines a sparse grid, providing geometric priors and structural guidance, with an implicit dense grid that enriches scene representation. By coupling the VDB structure with a submap-based organization, the framework reduces computational load and enables efficient incremental mapping on a large scale. To mitigate discontinuities across submaps, we introduce a distillation-based overlap alignment strategy, in which preceding submaps supervise subsequent ones to ensure consistency in overlapping regions. To further enhance robustness and sampling efficiency, we incorporate a dynamic removal module. Extensive experiments show that our approach delivers superior mapping quality while overcoming the efficiency limitations of voxel-guided methods, thereby outperforming existing state-of-the-art mapping methods.

Problem

• Addresses inefficiency in neural LiDAR mapping for large-scale environments.
• Combines explicit and implicit grid representations to enhance mapping performance.
• Improves real-time incremental mapping with submap alignment and dynamic removal.

Innovation

• Hybrid explicit-implicit grid for neural LiDAR mapping
• VDB-based submaps with distillation alignment for consistency
• Dynamic removal module enhances robustness and sampling efficiency