Attention-Based Neural-Augmented Kalman Filter for Legged Robot State Estimation

Seokju Lee, Kyung-Soo Kim · 2026

Read Attention-Based Neural-Augmented Kalman Filter for Legged Robot State Estimation by Seokju Lee, Kyung-Soo Kim (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-18569.txt.

Description

Seokju Lee and Kyung-Soo Kim give legged robots a better way to know where they stand and how they move. A legged robot reads its pose from joint cues and from a chip that feels push and turn. When a foot slips, the joint cues lie: they claim the foot held firm when it slid. That lie bends the pose read and can make the robot fall.

Their tool blends the old maths of such reads with a smart net that keeps watch for slip. The net learns which cues to trust at each tick, and it flags the size of the slip lie. The read can then fix the lie as it goes. A firm read is key for robots that walk on wet, loose, or rough ground. Such robots could work in fields, plants, and quake zones.

Catalog record

CREATORSeokju Lee, Kyung-Soo Kim
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE42 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-18569

Browse the full library · About the Tower of Babel