ETH and NASA JPL open-source Holistic Fusion: task-agnostic robot localization via factor graphs
rsasaki0109 · x · 2026-10-15
Holistic Fusion (HF) is a robot localization and state-estimation paper published in IEEE T-RO 2026, from ETH Zürich's Robotic Systems Lab, the Max Planck Institute for Intelligent Systems, and NASA JPL.
Key ideas
- Fuses IMU, absolute, landmark, and local measurements from an arbitrary number of sensors and reference frames in one factor graph.
- Jointly estimates local and global robot state plus a theoretically unlimited number of dynamic variables with automatic reference-frame alignment — no task-specific hard-coding needed.
- Measurements expressed in different frames are included explicitly as states, with their evolution modeled as random walks.
Resources
Open-sourced with code, datasets, docs, and examples; real-world deployments cover the ANYmal quadruped, the RACER off-road vehicle, and the HEAP walking excavator.
Related event: ETH and NASA JPL open-source Holistic Fusion for robot localization(2 posts)→
More from Embodied
- Scoble: Tesla FSD 14.3 update is spectacular even on 8-year-old Hardware 3 cars — Scobleizer · 2026-10-15
- Dev puts a Muse AI agent on the wall with Pi 5 and laser projector — alexandr_wang · 2026-10-15
- NavGPT-3 Reaches Human-Level Navigation: 90.43 SR on RxR-CE vs Human 90.4 — Metacognition-AI · 2026-10-15
- Figure 02 humanoid crosses obstacles blind, powered by an RL controller — jonstephens85 · 2026-10-15
- HapticWorld: A world simulator that lets teleoperators feel joint torque in real time — Scobleizer · 2026-10-15
- Rice's RoboTok Talk: Mining Internet Videos for Dexterous Robot Learning — ceciletamura · 2026-10-15