arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.11972cs.ROcs.LG
收藏

通过图神经网络生成因子化3D场景图中的不确定性感知高层空间概念

Generation of Uncertainty-Aware High-Level Spatial Concepts in Factorized 3D Scene Graphs via Graph Neural Networks

Automation and Robotics Research Group, Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg(卢森堡大学自动化与机器人研究组,安全、可靠性与信任跨学科研究中心(SnT))

Jose Andres Millan-Romera, Muhammad Shaheer, Miguel Fernandez-Cortizas, Martin R. Oswald, Holger Voos, Jose Luis Sanchez-Lopez

AI总结:

本文提出一种基于学习的方法,通过在线推断垂直平面中的空间概念,并将其作为SLAM后端可优化的因素,从而提升室内导航和映射的鲁棒性。

摘要:

使机器人能够从原始几何观测(如平面表面)中自主发现高层空间概念(如房间和墙壁)对于可靠的室内导航和建图至关重要。这些图提供了一种分层的度量-语义表示,其中这些概念被组织起来。为了进一步提高图SLAM性能,因子化3D场景图将这些概念作为优化因素,这些因素约束相对几何并强制全局一致性。然而,这两个阶段仍然主要依赖人工:概念通常通过手工制作的、特定概念的启发式方法推导出来,而因素及其协方差同样手动设计。这种对人工规范的依赖限制了在多样化环境中的泛化能力和对新概念类别的扩展能力。本文提出了一种新的基于学习的方法,从观察到的垂直平面中在线推断空间概念,并将其引入SLAM后端作为可优化的因素,从而消除了手工制作概念生成、因素设计和协方差规范的需要。我们在具有复杂布局的模拟环境中评估了我们的方法,提高了房间检测的20.7%和轨迹估计的19.2%。在真实施工场地上验证,房间检测提高了5.3%,地图匹配精度提高了3.8%。

英文摘要:

Enabling robots to autonomously discover high-level spatial concepts (e.g., rooms and walls) from primitive geometric observations (e.g., planar surfaces) within 3D Scene Graphs is essential for robust indoor navigation and mapping. These graphs provide a hierarchical metric-semantic representation in which such concepts are organized. To further enhance graph-SLAM performance, Factorized 3D Scene Graphs incorporate these concepts as optimization factors that constrain relative geometry and enforce global consistency. However, both stages of this process remain largely manual: concepts are typically derived using hand-crafted, concept-specific heuristics, while factors and their covariances are likewise manually designed. This reliance on manual specification limits generalization across diverse environments and scalability to new concept classes. This paper presents a novel learning-based method that infers spatial concepts online from observed vertical planes and introduces them as optimizable factors within a SLAM backend, eliminating the need to handcraft concept generation, factor design, and covariance specification. We evaluate our approach in simulated environments with complex layouts, improving room detection by 20.7% and trajectory estimation by 19.2%. Validated on real construction sites, room detection improves by 5.3% and map matching accuracy by 3.8%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑