Functional-SLAM: First SLAM Framework to Maintain Functional Scene Graphs Online for Robotic Interaction

zhenjun_zhao · x · 2026-09-11

Tsinghua researchers present Functional-SLAM, the first framework that continuously maintains a functional scene graph as an online SLAM state.

• Existing SLAM systems lack the functional relation modeling needed for fine-grained robotic interaction, while prior functional 3D scene graph methods rely on offline reconstruction
• The method combines anchor-keyframe geometry with functional-context constraints for persistent nodes, accumulates multi-frame evidence for stable functional edges, and uses functional topology to supplement visual loop closure in repetitive or texture-degraded scenes
• Experiments show stable online functional mapping with much faster runtime than offline methods, plus improved pose accuracy from topology-assisted loop closure
• Code is publicly available

Original post →

More from Embodied

• LIBERO-MAX: Robot Policies Drop 11–26 Points When the World Changes Mid-Task — drmapavone · 2026-10-27
• Photos from ROSCon UK: BBC's robotic whale, da Vinci surgical robot, PR2 and Baxter — kamathsblog · 2026-10-27
• Agility's Digit robot to expand into two more GXO US warehouses from early 2027 — jonstephens85 · 2026-10-27
• Real2Sem2Real turns one phone video into 2,000+ robot demos, matching 150 teleop demos — Scobleizer · 2026-10-27
• Wayve CEO argues end-to-end deep learning should power driving, safety and world models — alexgkendall · 2026-10-27
• Mundo turns a few room photos into an explorable, simulatable 3D world — at 1.5 FPS — mark_k · 2026-10-27

Embodied channel →