arXiv cs.RO·· 2 days agoAI score62
MosaicIMU: A Generalizable Neural Inertial Odometry Framework for Multiple Carriers
Summary
MosaicIMU is a carrier-conditioned Mixture-of-Experts (MoE) pretraining and adaptation framework for generalizable neural inertial odometry. It composes carrier-specific expert features through a prototype-based router, decodes local velocity and uncertainty constraints, and integrates them with a history-aware EKF. For unseen domains, it freezes the pretrained base model and learns a lightweight expert residual branch; for edge deployment, it reuses the router to select online samples for incremental updates.
This page provides a summary and source link only. Read the full article at the source.
Source: arXiv cs.RO · arxiv.org
Perception and localizationRobot models and learningResearch papers#Research#Perception and localization#Robot models and learning
Part of the event: MosaicIMU: Carrier-Conditioned MoE Neural Inertial Odometry