Search papers, labs, and topics across Lattice.
This paper introduces MosaicIMU, a Mixture-of-Experts (MoE) framework designed to enhance the generalization of neural inertial odometry across diverse carriers. By leveraging a prototype-based router to combine carrier-specific expert features and integrating them with a history-aware Extended Kalman Filter (EKF), MosaicIMU effectively reduces integration drift and improves performance in unseen environments. Experimental results demonstrate a significant reduction in average Absolute Trajectory Error (ATE) and Relative Trajectory Error over 10 seconds (RTE-10s) by 40% and 34%, respectively, compared to existing learning-based methods.
MosaicIMU achieves a 40% reduction in trajectory error by seamlessly adapting inertial odometry across different carriers using a novel Mixture-of-Experts approach.
Robust inertial odometry is essential for various carriers when external sensing is unreliable. Learning-based methods reduce integration drift by capturing local motion priors, but these methods often remain tied to a particular carrier, limiting generalization across heterogeneous platforms. We present MosaicIMU, a carrier-conditioned Mixture-of-Experts (MoE) pretraining-and-adaptation framework for generalizable neural inertial odometry. MosaicIMU uses a prototype-based router to compose carrier-specific expert features, decodes local velocity and uncertainty constraints, and integrates them with a history-aware EKF. For unseen domain adaptation, it freezes the pretrained base model and learns a new lightweight expert residual branch. For edge-deployment, it further reuses the router to select informative online samples for efficient incremental updates. Experiments show that MosaicIMU consistently outperforms learning-based baselines, reducing average ATE and RTE-10s by 40% and 34%, respectively. These results highlight that MosaicIMU provides a scalable pretraining-to-deployment paradigm for generalizable and adaptive neural inertial odometry.
